…l-closed private_paths, bounded secret extractor

Resolves four review findings on the coasts-borrowed env features:

- #1 (High): service declarations are now pinned at `env create` into an
  env-local services.json (immutable from the box) with a service_digest in the
  manifest; service_start loads the pinned manifest and verifies the digest
  (fail-closed), so editing the worktree .h5i/env.toml after create can't change
  what a service runs. The start event records the redacted command.
- #2 (High): `env ports` reframed as port INJECTION, not forwarding — there is
  no host->box forwarder in v1, so the URL is rendered conditional ('if the
  service binds the injected port'), never a guarantee; CLI/MCP/docs updated.
- #3 (Medium): comma-bearing private_paths now fail closed — rejected in
  validate_private_paths (rel) and prepare_private_paths (container backing)
  instead of being silently skipped in build_run_argv.
- #4 (Medium): the opted-in command: secret extractor is now bounded — 10s wall
  timeout + 1 MiB stdout cap, own process group (killpg on timeout), drain-to-EOF
  reader (no pipe deadlock, bounded memory), all fail-closed.

Also updates the MCP tool-count test for the env tools added earlier.

Tests: lib 891 pass; env_integration 98 pass. Clippy clean on default and
--no-default-features.

…mon-free services + dynamic ports (#169)

* Add docs/borrowing-from-coasts.md: ideas to improve h5i env from Coasts

Survey of ../coasts/ (runtime orchestrator: ports, secrets keystore, daemon+UI,
agent-shells, private_paths, shared services) mapped onto h5i env's
confinement+provenance core. Nine borrow ideas with posture guards (no mandatory
daemon, isolated-by-default, in-box agent). Codex consulted via i5h over two
rounds; governing principle: every borrowed affordance = policy grant + event +
capture, or don't ship it.

* env: fleet/doctor, private_paths, gated secret command extractor, daemon-free services + dynamic ports

Implements ideas borrowed from ../coasts (docs/borrowing-from-coasts.md) for
`h5i env`, each with CLI + MCP parity and tests:

- Idea 0 (fleet/doctor): `env list --json` (manifest + drift), new `env doctor`
  with a per-env enforcement-readiness check (functional verify_exec self-test)
  plus policy-digest/branch/context/drift health; exits non-zero when unhealthy.
- Idea 3 (private_paths): new [profile.X.private_paths] policy field
  (kind/persist); per-env backing dirs bind-mounted over workspace paths to give
  concurrent same-repo envs distinct inodes (no cross-env flock/cache
  contention) — pre_exec binds on process/supervised, --mount on container;
  fail-closed validation (relative, no .., no overlap, kind allowlist).
- Idea 1 (secrets): gated command: extractor (allow_command_extractors, pinned
  in the policy digest, refused at load and resolve) + `env secrets` legibility
  CLI (status + fingerprint, never values).
- Idea 3.5 (services): daemon-free `env service start|stop|status|logs`
  (sandbox::spawn_background, pid registry, logs-as-captures on stop, start/stop
  events on refs/h5i/env).
- Idea 2 (ports): per-service dynamic host port allocated + injected
  (H5I_ENV_PORT_<NAME>/PORT), surfaced by `env ports`.

Audit note: the flagged exec patterns are intentional — the confined background
spawn (pre_exec) and the policy-gated command: extractor (sh -c).

Tests: env_integration 96 pass; lib sandbox/secrets/container 80 pass;
sandbox_api 4 pass. Clippy clean on default and --no-default-features.

* env: address review — pin service defs, port-injection semantics, fail-closed private_paths, bounded secret extractor

Resolves four review findings on the coasts-borrowed env features:

- #1 (High): service declarations are now pinned at `env create` into an
  env-local services.json (immutable from the box) with a service_digest in the
  manifest; service_start loads the pinned manifest and verifies the digest
  (fail-closed), so editing the worktree .h5i/env.toml after create can't change
  what a service runs. The start event records the redacted command.
- #2 (High): `env ports` reframed as port INJECTION, not forwarding — there is
  no host->box forwarder in v1, so the URL is rendered conditional ('if the
  service binds the injected port'), never a guarantee; CLI/MCP/docs updated.
- #3 (Medium): comma-bearing private_paths now fail closed — rejected in
  validate_private_paths (rel) and prepare_private_paths (container backing)
  instead of being silently skipped in build_run_argv.
- #4 (Medium): the opted-in command: secret extractor is now bounded — 10s wall
  timeout + 1 MiB stdout cap, own process group (killpg on timeout), drain-to-EOF
  reader (no pipe deadlock, bounded memory), all fail-closed.

Also updates the MCP tool-count test for the env tools added earlier.

Tests: lib 891 pass; env_integration 98 pass. Clippy clean on default and
--no-default-features.

* env: close service-pinning gaps from re-review (no-service envs, name validation)

Two High findings from Codex's re-review of the service feature:

- No-service envs were left with service_digest=None, which routes service_start
  to the legacy fallback that reads the mutable worktree/repo .h5i/env.toml — so
  an env created with no services could add [service.*] afterwards and start it
  unpinned. Fix: pin_services_at_create now ALWAYS writes services.json and
  records a digest (even for the empty set), so every new env is pinned (empty or
  not) and the mutable fallback only applies to truly pre-pinning legacy envs.

- Service names were used directly in services/<name>.json and <name>.log paths,
  so a key like ../manifest could escape services/ and overwrite env-local files.
  Fix: validate_service_name (strict slug — rejects '/', '..', leading-dot)
  enforced at parse/pin time (create fails closed on a bad [service.*] key) AND
  at the top of service_start/stop/logs before any path is built.

Tests: no-service env can't start an added service; traversing names rejected at
start and at create. env_integration 101 pass; clippy clean.

* docs: document env fleet/doctor, secrets, services, ports + private_paths

Update MANUAL.md, the man page (man/man1/h5i.1), and the regenerated
docs/manual/index.html for the new env surface:

- Command table / man subcommands: env list --json, env doctor, env secrets,
  env service start|stop|status|logs, env ports; service added to the event list.
- Policy file: [profile.X.private_paths] (kind/persist), the gated command:
  secret source + allow_command_extractors, and [service.<name>] (command/port/
  restart/logs), with notes that service defs are pinned at create.
- New 'Services and dynamic ports' subsection (TOC + anchor) covering pinning,
  confinement, logs-as-evidence, and injection-not-forwarding port semantics.
- Secrets-broker section: command: source (bounded, gated) + env secrets dry-run.

docs/manual/index.html regenerated via scripts/gen_manual.py.

Add /feed.xml (RSS 2.0, 27 posts newest-first) and a
<link rel="alternate" type="application/rss+xml"> in the head of the
homepage, blog index, and all 27 posts so feed readers and answer engines
can discover the feed.

* docs/blog: improve 12 posts per review (workflow clarity, hook docs, examples)

Address reviewer feedback across the blog cluster, grounded in the actual
CLI (verified h5i hook surface + human-vs-agent command split):

- structured-tool-output: add raw verbose pytest output before the compact
  capture so the ~95% compression is visible.
- token-reduction: clarify h5i capture run is agent-invoked (PreToolUse
  wrap-bash rewrite), not hand-typed.
- persistent-memory, track-prompts, context-dag: make the human-vs-agent
  split explicit (human wires hooks once; handlers fire automatically);
  context-dag's outdated Claude/Codex hook 'daily workflow' rewritten.
- claude-code-hooks-vs-git-hooks: add h5i hook setup code block + option
  and handler/event reference tables + Codex note.
- uncertainty-heatmap, auditing-ai-generated-code: add h5i hook setup as
  the first getting-started step.
- prompt-maturity-score: drop the redundant 'Honesty about what this is'
  section.
- why-git-diffs, what-is-ai-aware: add concrete git-vs-h5i usage blocks.
- review-code-written-by-ai-agents: cross-link each section to its post.

Fixed bare 'h5i share pr' -> 'h5i share pr post'. All posts HTML-balanced.

* docs/blog: fix wrong claim that Codex lacks the wrap-bash PreToolUse hook

merge_codex_config_toml (+ codex_config_toml_wrap_bash_is_idempotent test)
shows h5i hook setup --write --target codex --wrap-bash writes the same
PreToolUse/Bash 'h5i hook wrap-bash' handler into .codex/config.toml.
Codex only lacks the PostToolUse/UserPromptSubmit core hooks (mined from
session JSONL); wrap-bash works for both runtimes.

- token-reduction: drop 'for agents without a PreToolUse hook, such as
  Codex, wrapped explicitly'; state the hook works for both (--target codex).
- claude-code-hooks-vs-git-hooks: note --wrap-bash is not Claude-only.

* docs/blog: deep-rewrite 6 shallow comparison/concept posts

Senior-blogger pass on the thinnest, highest-SEO posts: deepen in place
(structure/URL/canonical/internal links preserved), ground every h5i
claim in source, add FAQPage JSON-LD, sharpen titles/meta, drop hype.

- git-notes-vs-h5i: reframe to 'h5i is built ON git notes' (record is
  JSON on refs/h5i/notes); add how-notes-work, multi-ref architecture
  (notes/context/msg/memory), sharing semantics, failure modes.
- git-communication-layer: add the CAS-send + union-merge-by-id mailbox
  mechanics, fair comparison vs chatops/bus/MCP/A2A, failure modes.
- what-is-ai-aware-version-control: reorient to the head term (fix H1),
  define the six-field record, fair comparison (trailers/PR templates/
  MLOps lineage/CI/vendor transcripts), worked example.
- why-git-diffs: concrete reviewer scenarios, volume!=risk, context
  coverage as signal, fair comparison incl. LLM-as-reviewer.
- claude-code-hooks-vs-git-hooks: two-lifecycle SEE-table, worked
  layering example, advisory-vs-enforced trust model.
- review-code: deepen each of 7 steps (technique/failure/command),
  worked review packet, human-vs-LLM-vs-provenance comparison.

Verified against source (refs/h5i/notes, append_message_cas, --show-prompt,
audit review, share pr post); no invented flags.

* docs/blog: deep-rewrite 10 feature deep-dive posts

Senior-blogger pass on the feature cluster (deepen in place, ground every
claim in source, add FAQPage JSON-LD, sharpen titles/meta, drop hype,
real conclusions). Fixed several pre-existing inaccuracies found against
source:

- persistent-memory: 'every capture commit snapshots memory' was wrong —
  that's the context workspace; memory is h5i capture memory (src/memory.rs).
- agent-ensembles: lifecycle phase names corrected to the real run.phase
  strings (draft/dispatched/sealed_submit/verified/discuss/verdict/applied).
- pr-body: 5 PR styles (added 'minimal'), default is receipt not review.
- token-reduction: removed a non-existent experiment script; re-anchored
  to tests/filter_quality.rs.
- from-git-blame: corrected invented --show-prompt output, dropped the
  non-existent integrity_report field, --prompt→--intent, fixed audit vibe.
- uncertainty-heatmap: removed 'AI's own calibration' overclaim (it's a
  hand-assigned lexical weight).

Each post: why-now opener, fair comparisons, failure modes, worked
examples, verified flags only. i5h reframed as the wire-format spec.

* docs/blog: deep-rewrite sandboxing series + pillar (6 posts)

Senior-blogger pass on the sandboxing cluster, grounded in src (sandbox.rs,
supervisor.rs, container.rs, env.rs); deepen in place, fair comparisons,
honest isolation claims, FAQPage JSON-LD, real conclusions, no hype.

- foundations (Part 1): 'bits present != confinement works' via verify_exec;
  honest side-channel / shared-kernel ceiling.
- implementation (Part 2): probe→verify→fail-closed; corrected memory rlimit
  to RLIMIT_DATA (src), Landlock HardRequirement, deny-list rationale,
  deferred seccomp-notify execve, AppArmor/CI failure modes.
- landscape (Part 3): fair 6-axis comparison (VM/microVM/gVisor/Kata/
  container/kernel-confinement/none); honest h5i placement (no VM tier, L7 egress).
- sandboxing-h5i (Part 4): tier ladder + policy model; corrected egress —
  supervised = L3/L4 nftables/slirp, container = L7 CONNECT proxy, process
  can't (sandbox.rs:514); .h5i/env.toml example.
- agent-sandbox-env: repositioned around the propose/apply review loop +
  mediated commit; triple fusion.
- auditable-workspaces (pillar): refs/hooks mapping, fair comparison, kept
  distinct from + cross-linked to what-is-ai-aware-version-control.

Verified all tier/Landlock/seccomp/Podman claims against source.

* docs: correct CLAUDE.md memory rlimit (RLIMIT_DATA, not RLIMIT_AS)

src/sandbox.rs:1669 caps memory with RLIMIT_DATA (RLIMIT_AS over-counts
V8/Go virtual reservations and would OOM legit builds at ~100MiB); the
sandbox.rs module note said 'AS'. Surfaced while deep-rewriting the
sandboxing-implementation blog post against source.

* docs/blog: deep-rewrite security/CVE cluster (4 posts)

- cve-2025-59536, cve-2026-33068: conservative pass — reframed openers,
  added FAQ + FAQPage schema, sharpened titles; ALL CVE facts (ids, CWE,
  CVSS, affected/fixed versions, dates, mechanism, remediation) preserved.
- prompt-injection-in-agent-traces: fixed major fabrication vs src — the
  rule table is 4 HIGH/3 MEDIUM/1 LOW (src/injection.rs), not the invented
  'eight rules'; removed invented audit-scan flags; grounded sanitize_display
  examples in the unit tests; honest residual-risk framing.
- auditing-ai-generated-code: corrected min_score 0.40→0.5, removed
  fabricated true-positive rates and non-existent flags; clarified only 2 of
  4 vectors feed audit review (uncertainty/injection are separate scans).

All verified against source; FAQPage JSON-LD added; no hype.

* docs: SEO/AEO — add llms.txt, refresh sitemap, welcome AI crawlers

- llms.txt: curated markdown index (docs + concepts/features/sandboxing/
  security blog clusters + guides) for answer engines / LLM browsing (AEO).
- sitemap.xml: set <lastmod> 2026-06-26 on all 27 blog posts (freshness
  signal after this session's rewrites); ADDED the orphaned
  agent-ensembles-with-h5i-team URL (was missing); non-blog pages untouched.
- robots.txt: explicit Allow stanzas for GPTBot/OAI-SearchBot/ClaudeBot/
  PerplexityBot/Google-Extended/etc. + llms.txt hint.
- bumped dateModified to 2026-06-26 on 4 deep-rewritten posts whose agents
  left it at 2026-06-03 (page now agrees with sitemap).

* docs: SEO/AEO #4 — add RSS feed + alternate link

Add /feed.xml (RSS 2.0, 27 posts newest-first) and a
<link rel="alternate" type="application/rss+xml"> in the head of the
homepage, blog index, and all 27 posts so feed readers and answer engines
can discover the feed.

* docs/blog: SEO/AEO #5 — add Key takeaways box to all 27 posts

Insert a styled 'Key takeaways' callout (3 bullets) at the top of every
blog post, right after the header. Bullets are the verified, post-specific
takeaways from this session's rewrites — a TL;DR answer block that gives
answer engines / featured snippets a concise extract and orients readers.
Reuses the existing .callout style (+ kt-takeaways marker).

* docs/blog: group flagship posts under a new Ensemble pillar

Make agent-ensembles + auditable-workspaces a named 'Ensemble' pillar (the
site's run-many-merge-one narrative):
- eyebrows -> 'Ensemble · h5i team' / 'Ensemble · Auditable workspaces'
- blog index: new lead '00 · Ensemble' post-group above the two flagship
  cards; relabel their card metas; reframe the index intro to lead with the
  Ensemble pillar (the other four become supporting pillars).
- cross-link the pair (auditable-workspaces now has a Read-next card to
  agent-ensembles; the reverse link already existed).
- llms.txt: new top-level Ensemble section holding both (de-duped from
  Features/Sandboxing).

…kage

Replaces v1's flat weighted sum + special-case caps with a slot rubric and a
multiplicative core, the shape the bug-report-quality (CTQRS) and code-gen
prompt-guideline (Midolo et al., arXiv 2601.13118) literature converge on.

Aggregation (#2, #3): three necessary CORE slots — objective (a positive,
actionable goal, not just prohibitions), grounding (concrete refs), direction
(acceptance + constraints + pre/post/IO/exception contract) — combine as a
floored geometric mean, so a prompt empty on any axis is capped with no
hand-coded gate. Six ENRICHMENT signals (context, examples, structure,
diversity, clarity, adequacy) lift an already-solid core additively. The
tactical exemption now falls out for free: context is enrichment, not core.
Retired apply_balance_gates (the 69/79 caps) and the tiered length caps.

New slot detectors (#2): behavioral-contract cluster (pre/post-conditions,
I/O format, exceptions), an Examples slot (markers + input->output arrows),
and executable-acceptance detection (a named runner like 
running 941 tests
test cgroup::tests::format_limit_renders_max_or_number ... ok
test cgroup::tests::parse_count_handles_max_and_numbers ... ok
test cgroup::tests::parse_cpu_usage_from_cpu_stat ... ok
test claude::tests::keyword_search_empty_commit_list ... ok
test claude::tests::keyword_search_is_case_insensitive ... ok
test claude::tests::keyword_search_returns_best_match ... ok
test claude::tests::keyword_search_returns_none_when_no_terms_match ... ok
test claude::tests::sanitize_handles_multiple_and_unclosed_reminders ... ok
test claude::tests::keyword_search_searches_prompt_field ... ok
test claude::tests::sanitize_does_not_match_longer_tag_name ... ok
test claude::tests::sanitize_drops_background_task_notification_turn ... ok
test claude::tests::keyword_search_higher_score_wins ... ok
test claude::tests::sanitize_keeps_plain_human_prompt ... ok
test claude::tests::sanitize_strips_system_reminder_but_keeps_human_text ... ok
test claude::tests::sanitize_drops_system_notification_banner_turn ... ok
test claude::tests::sanitize_system_reminder_only_is_dropped ... ok
test cgroup::tests::probe_is_honest_about_this_host ... ok
test claude::tests::sanitize_trims_whitespace ... ok
test cli_routing::tests::known_noun_verb_rewrites_to_legacy_and_keeps_rest ... ok
test cli_routing::tests::levenshtein_basic ... ok
test cli_routing::tests::missing_or_help_verb_requests_help ... ok
test cli_routing::tests::nearest_verb_respects_distance_cap ... ok
test cli_routing::tests::refspec_metacharacters_are_rejected ... ok
test cli_routing::tests::scoped_context_push_targets_single_ref ... ok
test cli_routing::tests::too_short_or_non_noun_passes_through ... ok
test cli_routing::tests::unscoped_context_push_uses_wildcard ... ok
test cli_routing::tests::unknown_verb_suggests_nearest ... ok
test cli_routing::tests::valid_branch_names_pass ... ok
test codex::tests::extract_human_prompt_ignores_synthetic_context_messages ... ok
test codex::tests::extract_human_prompt_reads_codex_user_message ... ok
test codex::tests::extract_patch_events_reports_file_actions ... ok
test cgroup::tests::live_scoped_cgroup_applies_and_accounts ... ok
test codex::tests::extract_events_custom_tool_call_ignores_other_names ... ok
test codex::tests::extract_events_exec_command_rg_files ... ok
test codex::tests::extract_events_reads_parsed_exec_commands ... ok
test codex::tests::extract_events_exec_command_sed_read ... ok
test codex::tests::extract_events_response_item_custom_tool_call_apply_patch ... ok
test codex::tests::extract_events_top_level_custom_tool_call_apply_patch ... ok
test codex::tests::shell_cmd_compile_ignored ... ok
test container::tests::allowlist_does_not_treat_ipv6_as_port ... ok
test container::tests::codex_profile_is_detected_for_managed_settings_gating ... ok
test container::tests::allowlist_exact_wildcard_and_port ... ok
test container::tests::empty_allowlist_denies_everything ... ok
test codex::tests::shell_cmd_rg_files_flag ... ok
test codex::tests::shell_cmd_pipeline_only_first_command ... ok
test codex::tests::shell_cmd_cat_multiple_files ... ok
test container::tests::egress_tally_clamps_to_max_hosts ... ok
test container::tests::egress_tally_counts_and_orders_verdicts ... ok
test container::tests::parse_target_connect_and_absolute ... ok
test codex::tests::shell_cmd_rg_search_with_path ... ok
test codex::tests::shell_cmd_sed_extracts_filename ... ok
test container::tests::run_argv_injects_brokered_secret_env ... ok
test container::tests::run_argv_mounts_env_inbox_read_only ... ok
test container::tests::run_argv_mounts_managed_settings_read_only ... ok
test container::tests::run_argv_mounts_env_capture_spool_without_shim ... ok
test container::tests::run_argv_mounts_box_git_plumbing_at_identical_paths ... ok
test container::tests::run_argv_interactive_adds_i_and_optional_t ... ok
test container::tests::run_argv_mounts_agent_hook_configs_read_only ... ok
test container::tests::run_argv_shim_mounts_image_shells_and_spool ... ok
test container::tests::run_argv_proxy_mode_sets_proxy_env ... ok
test container::tests::run_argv_is_hardened ... ok
test container::tests::run_argv_mounts_private_paths_rw_under_work ... ok
test container::tests::shim_script_tees_and_passes_through ... ok
test container::tests::proxy_gates_connect_by_allowlist ... ok
test ctx::tests::append_log_bootstraps_context_ref_without_init ... ok
test ctx::tests::append_log_bootstrap_trace_is_readable ... ok
test ctx::tests::context_diff_fails_when_snapshot_missing ... ok
test ctx::tests::append_log_adds_entry_to_trace ... ok
test ctx::tests::append_log_uppercases_kind ... ok
test ctx::tests::current_branch_defaults_to_main ... ok
test ctx::tests::context_diff_detects_added_trace_lines ... ok
test ctx::tests::dag_head_id_empty_when_dag_is_empty ... ok
test ctx::tests::context_diff_same_snapshot_shows_no_changes ... ok
test ctx::tests::context_diff_detects_added_milestones ... ok
test ctx::tests::auto_follow_switches_ctx_branch_to_match_git_branch ... ok
test ctx::tests::dag_head_id_matches_last_node ... ok
test ctx::tests::context_diff_detects_new_trace_lines_after_pack ... ok
test ctx::tests::context_diff_uses_snapshot_branch_not_current_branch ... ok
test ctx::tests::dag_is_empty_after_init ... ok
test ctx::tests::dag_links_parent_ids_in_chain ... ok
test ctx::tests::dag_records_node_on_append_log ... ok
test ctx::tests::dag_merge_node_has_two_parents ... ok
test ctx::tests::distill_knowledge_deduplicates_across_branches ... ok
test ctx::tests::dag_node_ids_are_unique ... ok
test ctx::tests::extract_list_items_parses_bullet_list ... ok
test ctx::tests::extract_recent_commits_returns_latest_first_when_multiple ... ok
test ctx::tests::extract_section_returns_correct_content ... ok
test ctx::tests::extract_section_returns_empty_when_missing ... ok
test ctx::tests::distill_knowledge_collects_think_entries ... ok
test ctx::tests::distill_knowledge_includes_thoughts_from_all_branches ... ok
test ctx::tests::ephemeral_trace_not_in_trace_md ... ok
test ctx::tests::ephemeral_trace_not_in_dag ... ok
test ctx::tests::ephemeral_cleared_on_context_commit ... ok
test ctx::tests::gcc_branch_stores_purpose_retrievable_from_commit_md ... ok
test ctx::tests::explicit_checkout_pins_against_auto_follow ... ok
test ctx::tests::ephemeral_trace_visible_in_ephemeral_md ... ok
test ctx::tests::gcc_branch_switches_active_branch ... ok
test ctx::tests::gcc_checkout_switches_to_existing_branch ... ok
test ctx::tests::ephemeral_trace_on_non_main_branch_is_isolated ... ok
test ctx::tests::gcc_checkout_fails_on_nonexistent_branch ... ok
test ctx::tests::gcc_commit_appends_entry ... ok
test ctx::tests::gcc_commit_multiple_entries_all_visible ... ok
test ctx::tests::gcc_commit_updates_main_md_notes ... ok
test ctx::tests::gcc_commit_with_empty_detail_succeeds ... ok
test ctx::tests::gcc_context_reads_goal_from_main_md ... ok
test ctx::tests::gcc_context_active_branches_includes_all ... ok
test ctx::tests::gcc_context_reads_milestones ... ok
test ctx::tests::gcc_merge_fails_for_nonexistent_branch ... ok
test ctx::tests::gcc_context_returns_recent_commits ... ok
test ctx::tests::gcc_context_includes_current_branch ... ok
test ctx::tests::gcc_context_window_limits_returned_commits ... ok
test ctx::tests::gcc_scope_creates_scope_prefixed_branch ... ok
test ctx::tests::goal_log_collapses_unchanged_reinit ... ok
test ctx::tests::gcc_merge_combines_branches ... ok
test ctx::tests::gcc_merge_returns_error_on_conflicting_changes ... ok
test ctx::tests::gcc_scope_is_identified_as_scope_branch ... ok
test ctx::tests::gcc_merge_produces_two_parent_commit_on_target_ref ... ok
test ctx::tests::gcc_scope_can_be_merged_back ... ok
test ctx::tests::goal_log_empty_for_unknown_branch ... ok
test ctx::tests::gcc_merge_unions_divergent_dag_nodes_without_conflict ... ok
test ctx::tests::init_creates_workspace ... ok
test ctx::tests::init_eager_create_respects_pin ... ok
test ctx::tests::init_eager_create_is_idempotent ... ok
test ctx::tests::limit_recent_milestones_keeps_newest_tail_and_total ... ok
test ctx::tests::init_embeds_goal_in_main_md ... ok
test ctx::tests::init_eager_creates_active_shadow_for_git_branch ... ok
test ctx::tests::is_initialized_false_before_init ... ok
test ctx::tests::init_idempotent_does_not_overwrite_goal ... ok
test ctx::tests::is_initialized_true_after_init ... ok
test ctx::tests::goal_log_skips_unchanged_commits_and_records_reverts ... ok
test ctx::tests::goal_log_tracks_history_newest_first ... ok
test ctx::tests::list_branches_after_init_has_main ... ok
test ctx::tests::lossless_pack_is_noop_on_empty_trace ... ok
test ctx::tests::lossless_pack_keeps_standalone_observe_not_subsumed ... ok
test ctx::tests::lossless_pack_merges_consecutive_observe ... ok
test ctx::tests::lossless_pack_preserves_note_entries_verbatim ... ok
test ctx::tests::lossless_pack_noop_when_no_observe_entries ... ok
test ctx::tests::list_branches_includes_new_branches ... ok
test ctx::tests::is_pin_misrouting_flags_stale_pin ... ok
test ctx::tests::pin_tracks_git_accepts_env_pairing_not_genuine_misroute ... ok
test ctx::tests::migrate_legacy_creates_per_branch_refs ... ok
test ctx::tests::lossless_pack_preserves_think_act_verbatim ... ok
test ctx::tests::pack_returns_zero_when_no_snapshots ... ok
test ctx::tests::mini_trace_populated_even_without_show_log ... ok
test ctx::tests::lossless_pack_removes_subsumed_observe ... ok
test ctx::tests::pack_returns_zero_when_history_already_compact ... ok
test ctx::tests::prepare_context_write_requires_current_git_branch_goal ... ok
test ctx::tests::print_status_surfaces_unreadable_repo_as_error ... ok
test ctx::tests::prepare_context_write_allows_multiple_context_branches_per_git_branch ... ok
test ctx::tests::prepare_context_write_requires_active_context_branch_purpose ... ok
test ctx::tests::relevant_finds_commit_mentions ... ok
test ctx::tests::relevant_finds_trace_mentions ... ok
test ctx::tests::relevant_returns_empty_when_no_mentions ... ok
test ctx::tests::reconcile_flags_merged_git_branch_with_unmerged_ctx_branch ... ok
test ctx::tests::relevant_finds_cross_branch_mentions ... ok
test ctx::tests::restore_fails_when_no_snapshot_exists ... ok
test ctx::tests::relevant_matches_by_file_basename ... ok
test ctx::tests::mini_trace_capped_at_eight_entries ... ok
test ctx::tests::restore_is_nondestructive_adds_new_commit ... ok
test ctx::tests::rm_branch_errors_for_missing_branch ... ok
test ctx::tests::rm_branch_refuses_env_owned_branch ... ok
test ctx::tests::rm_branch_refuses_active_without_force_then_force_resets_head ... ok
test ctx::tests::rm_branch_refuses_main ... ok
test ctx::tests::restore_returns_ok_for_existing_snapshot ... ok
test ctx::tests::rm_branch_removes_an_inactive_branch ... ok
test ctx::tests::rm_branch_preserves_context_snapshots ... ok
test ctx::tests::search_returns_empty_when_no_match ... ok
test ctx::tests::search_finds_matching_trace_entry_with_file_mention ... ok
test ctx::tests::snapshot_for_commit_silently_skips_uninitialized_workspace ... ok
test ctx::tests::search_ranks_think_entries_higher ... ok
test ctx::tests::show_log_false_returns_empty_recent_log_lines ... ok
test ctx::tests::smart_recall_returns_empty_for_blank_query ... ok
test ctx::tests::smart_recall_ranks_task_specific_context ... ok
test ctx::tests::smart_recall_respects_limit_and_keeps_query ... ok
test ctx::tests::snapshot_for_commit_records_current_branch ... ok
test ctx::tests::stable_line_count_consistent_with_dynamic ... ok
test ctx::tests::snapshot_for_commit_writes_snapshot_file ... ok
test ctx::tests::union_dag_json_is_set_commutative_and_idempotent ... ok
test ctx::tests::union_dag_json_tolerates_empty_or_garbage_sides ... ok
test ctx::tests::union_dag_json_unions_by_id_and_dedups ... ok
test env::tests::agent_validation_blocks_traversal_and_bad_refs ... ok
test ctx::tests::todo_items_empty_when_no_keywords ... ok
test ctx::tests::todo_items_extracted_from_note_with_todo_keyword ... ok
test ctx::tests::stable_line_count_zero_when_trace_is_short ... ok
test env::tests::box_git_grants_cover_worktree_plumbing_and_nothing_protected ... ok
test ctx::tests::todo_items_extracted_from_think_with_fixme ... ok
test ctx::tests::todo_items_blocked_and_remaining_keywords_matched ... ok
test env::tests::conflict_runbook_points_at_in_box_resolution ... ok
test env::tests::context_snapshot_spool_is_named_so_ingest_never_drains_it ... ok
test env::tests::event_serde_roundtrip ... ok
test env::tests::build_env_provenance_caps_captures_and_counts_lanes ... ok
test ctx::tests::todo_items_not_extracted_from_observe_or_act ... ok
test env::tests::imported_manifest_validation_rejects_traversal_and_identity_tampering ... ok
test env::tests::manifest_serde_roundtrip ... ok
test env::tests::env_inbox_write_read_and_cursor_roundtrip ... ok
test env::tests::prepare_cargo_env_keeps_target_outputs_inside_worktree ... ok
test env::tests::commit_box_worktree_snapshots_untracked_edits ... ok
test env::tests::prepare_home_state_is_noop_at_workspace_tier ... ok
test env::tests::find_disambiguates ... ok
test env::tests::prepare_home_state_is_noop_for_non_agent_profiles ... ok
test env::tests::prepare_home_state_persists_and_does_not_reseed ... ok
test env::tests::materialize_persona_concatenates_excludes_and_digests ... ok
test env::tests::grant_box_git_applies_per_backend ... ok
test env::tests::remove_jsonl_drops_only_the_matching_id ... ok
test env::tests::resolve_work_rcfile_accepts_in_tree_and_rejects_escapes ... ok
test env::tests::prepare_private_tmp_redirects_shared_tmp_to_env_backing ... ok
test env::tests::prepare_home_state_redirects_agent_creds_to_per_env_copy ... ok
test env::tests::prepare_home_state_seed_prunes_bloat_keeps_credentials ... ok
test env::tests::prepare_home_state_skips_missing_paths_and_keeps_direct_grant ... ok
test env::tests::grant_box_git_reads_main_repo_project_config ... ok
test env::tests::read_staged_capture_round_trips_and_rejects_unsafe_ids ... ok
test env::tests::slug_validation ... ok
test env::tests::scan_nested_git_exempts_registered_submodules_only ... ok
test env::tests::staged_path_filter_handles_symlinks ... ok
test env::tests::staged_path_filter_rejects_the_dangerous_shapes ... ok
test env::tests::submitted_round_sentinel_is_monotonic ... ok
test env::tests::review_ops_refuse_while_run_lock_is_held ... ok
test env::tests::snapshot_for_submit_is_best_effort_under_run_lock ... ok
test env::tests::write_note_spool_sanitizes_filename_and_rejects_empty_oid ... ok
test env::tests::upsert_jsonl_replaces_by_id_and_keeps_others_sorted ... ok
test env::tests::write_team_review_spool_records_scoped_request ... ok
test env::tests::write_plain_bashrc_is_self_contained_and_skips_host_bashrc ... ok
test env::tests::scoped_merge_none_when_no_env_for_branch_and_no_base ... ok
test env::tests::scoped_code_branch_refs_lists_only_matching_envs ... ok
test env::tests::write_team_submit_spool_records_scoped_request ... ok
test filter_rules::tests::command_string_normalizes_argv0_basename ... ok
test env::tests::scoped_merge_keeps_only_envs_forked_from_branch ... ok
test env::tests::scoped_merge_preserves_envs_already_on_base ... ok
test ctx::tests::stable_line_count_with_exactly_40_entries ... ok
test hooks::tests::codex_config_toml_adds_core_hooks_and_preserves_settings ... ok
test hooks::tests::codex_config_toml_wrap_bash_is_idempotent ... ok
test hooks::tests::command_match_is_exact_or_space_delimited ... ok
test hooks::tests::default_leaves_existing_wrap_bash_alone ... ok
test hooks::tests::fresh_default_has_core_hooks_but_no_wrap_bash ... ok
test hooks::tests::idempotent_under_reapplication ... ok
test hooks::tests::legacy_agent_command_paths_are_migrated_in_place ... ok
test hooks::tests::legacy_codex_stop_hook_is_migrated_in_place ... ok
test hooks::tests::legacy_observe_bash_entry_is_always_stripped ... ok
test hooks::tests::managed_settings_carries_only_wrap_bash ... ok
test hooks::tests::preserves_unrelated_settings_and_msg_hook ... ok
test hooks::tests::rejects_non_object_settings ... ok
test hooks::tests::team_flag_adds_team_stop_hook_claude ... ok
test hooks::tests::team_flag_adds_team_stop_hook_codex ... ok
test hooks::tests::wrap_bash_flag_adds_pretooluse_bash_entry ... ok
test hooks::tests::wrap_shell_syntax_goes_through_bash_c ... ok
test hooks::tests::wrap_simple_command_keeps_real_argv ... ok
test hooks::tests::wrap_skips_h5i_cd_and_empty ... ok
test ctx::tests::stable_line_count_reflects_tail_boundary ... ok
test injection::tests::benign_text_is_clean ... ok
test injection::tests::known_injection_patterns_flag_expected_rules ... ok
test injection::tests::rule_description_resolves_known_rules_and_unknown_fallback ... ok
test injection::tests::more_hits_raise_score_until_capped ... ok
test injection::tests::severity_labels_and_weights_are_ordered ... ok
test injection::tests::scan_many_aggregates_like_newline_concatenation ... ok
test lfs::tests::batch_request_serializes_to_lfs_shape ... ok
test lfs::tests::batch_response_parses_actions_and_errors ... ok
test lfs::tests::basic_auth_only_leaks_to_same_origin ... ok
test lfs::tests::endpoint_derivation ... ok
test lfs::tests::only_404_501_are_unsupported ... ok
test lfs::tests::same_origin_compares_scheme_host_port ... ok
test lfs::tests::split_url_parts ... ok
test mcp::tests::blame_missing_file_param_is_error ... ok
test mcp::tests::capture_run_tool_passes_small_output_through_unstored ... ok
test mcp::tests::capture_run_tool_rejects_non_array_command ... ok
test filter_rules::tests::builtin_rules_all_parse ... ok
test filter_rules::tests::known_commands_route_to_expected_rules ... ok
test filter_rules::tests::registry_is_name_sorted_and_unambiguous ... ok
test filter_rules::tests::json_and_info_subcommands_fall_through ... ok
test filter_rules::tests::matches_and_filters_a_known_command ... ok
test filter_rules::tests::unknown_command_returns_none ... ok
test filter_rules::tests::builtin_golden_tests_pass ... ok
test mcp::tests::context_branch_auto_inits ... ok
test mcp::tests::context_branch_missing_name_is_error ... ok
test mcp::tests::context_checkout_missing_name_is_error ... ok
test mcp::tests::context_commit_missing_summary_is_error ... ok
test mcp::tests::context_branch_creates_branch_in_workspace ... ok
test mcp::tests::context_checkout_switches_active_branch ... ok
test mcp::tests::context_commit_auto_inits_and_records_summary ... ok
test mcp::tests::context_merge_missing_branch_is_error ... ok
test mcp::tests::context_show_not_initialized_returns_message ... ok
test mcp::tests::context_init_creates_workspace ... ok
test mcp::tests::context_status_not_initialized_returns_message ... ok
test mcp::tests::context_init_without_goal_still_succeeds ... ok
test mcp::tests::context_commit_uses_detail_when_provided ... ok
test mcp::tests::context_trace_missing_content_is_error ... ok
test mcp::tests::context_status_after_init_returns_structured_json ... ok
test mcp::tests::env_tools_are_advertised ... ok
test mcp::tests::context_trace_auto_inits_and_echoes_entry ... ok
test mcp::tests::context_trace_missing_kind_is_error ... ok
test mcp::tests::every_tool_has_description_and_object_schema ... ok
test mcp::tests::context_show_returns_gcc_context_json ... ok
test mcp::tests::initialize_returns_protocol_version_and_capabilities ... ok
test mcp::tests::initialize_advertises_subscribe_capability ... ok
test mcp::tests::notifications_return_no_response ... ok
test mcp::tests::env_run_unknown_env_errors ... ok
test mcp::tests::parse_error_on_invalid_json ... ok
test mcp::tests::resource_context_current_not_initialized_returns_false_flag ... ok
test mcp::tests::ping_returns_empty_object ... ok
test mcp::tests::resource_read_missing_uri_returns_32602 ... ok
test mcp::tests::resource_read_unknown_uri_returns_32603 ... ok
test mcp::tests::resource_context_current_initialized_returns_gcc_context ... ok
test mcp::tests::resource_snapshot_returns_empty_for_unknown_uri ... ok
test mcp::tests::response_err_sets_error_clears_result ... ok
test mcp::tests::response_ok_sets_result_clears_error ... ok
test mcp::tests::response_serialization_omits_null_fields ... ok
test mcp::tests::resource_snapshot_returns_nonempty_for_known_uris ... ok
test mcp::tests::resource_unknown_uri_is_error ... ok
test mcp::tests::resources_list_includes_context_and_log ... ok
test mcp::tests::full_context_workflow ... ok
test mcp::tests::tool_call_missing_name_returns_32602 ... ok
test mcp::tests::subscribe_known_uri_returns_empty_ok ... ok
test mcp::tests::subscribe_idempotent_on_second_call ... ok
test memory::tests::claude_memory_dir_encodes_path_separators ... ok
test memory::tests::claude_memory_dir_ends_with_memory ... ok
test mcp::tests::tool_call_unknown_tool_returns_is_error ... ok
test memory::tests::compute_diff_identical_returns_empty ... ok
test memory::tests::compute_diff_shows_context_around_change ... ok
test memory::tests::default_memory_dir_uses_codex_path_for_codex ... ok
test mcp::tests::tools_list_includes_all_expected_tools ... ok
test mcp::tests::unknown_method_returns_method_not_found ... ok
test memory::tests::codex_memory_dir_ends_with_memories ... ok
test memory::tests::diff_snapshots_error_on_missing_snapshot ... ok
test memory::tests::diff_snapshots_detects_removed_file ... ok
test memory::tests::lcs_diff_add_one_line ... ok
test memory::tests::lcs_diff_completely_different ... ok
test memory::tests::lcs_diff_empty_inputs ... ok
test memory::tests::lcs_diff_identical_is_all_context ... ok
test memory::tests::diff_snapshots_detects_added_file ... ok
test memory::tests::diff_snapshots_no_changes_when_identical ... ok
test memory::tests::diff_snapshots_detects_modified_file ... ok
test memory::tests::lcs_diff_remove_one_line ... ok
test memory::tests::list_snapshots_empty_when_no_memory_dir ... ok
test memory::tests::restore_snapshot_error_for_missing_oid ... ok
test mcp::tests::unsubscribe_removes_uri_from_map ... ok
test memory::tests::list_snapshots_returns_sorted_by_timestamp ... ok
test memory::tests::take_snapshot_copies_files ... ok
test metadata::tests::pending_context_without_human_prompt_field_parses ... ok
test metadata::tests::test_into_metrics_computes_total_from_components ... ok
test metadata::tests::test_into_metrics_defaults_all_zeroes ... ok
test metadata::tests::test_into_metrics_explicit_total_wins ... ok
test metadata::tests::test_into_metrics_propagates_tool_and_summary ... ok
test metadata::tests::test_is_passing_exit_code_zero ... ok
test metadata::tests::test_is_passing_legacy_coverage_fallback ... ok
test metadata::tests::test_is_passing_legacy_zero_coverage_fails ... ok
test metadata::tests::test_is_passing_no_exit_code_no_failures ... ok
test metadata::tests::test_is_passing_no_exit_code_with_failures ... ok
test metadata::tests::test_is_passing_exit_code_nonzero ... ok
test memory::tests::restore_snapshot_returns_correct_count ... ok
test memory::tests::take_snapshot_missing_source_creates_empty_snapshot ... ok
test metadata::tests::test_minimal_from_git_root_commit ... ok
test metadata::tests::test_minimal_from_git_child_commit ... ok
test msg::tests::broadcast_reaches_everyone_but_sender ... ok
test metadata::tests::test_timestamp_conversion_precision ... ok
test mcp::tests::env_compare_over_mcp ... ok
test msg::tests::empty_branch_opts_out_of_auto_tag ... ok
test msg::tests::empty_tag_is_normalised_to_none ... ok
test msg::tests::explicit_branch_overrides_auto_tag ... ok
test msg::tests::get_message_finds_by_id ... ok
test msg::tests::identity_resolution_prefers_explicit_then_stored ... ok
test msg::tests::infer_kind_passes_through_explicit_kind_names ... ok
test msg::tests::identity_resolution_refuses_shared_stored_in_multi_agent_clone ... ok
test msg::tests::identity_resolution_errors_when_unset ... ok
test msg::tests::legacy_v0_line_deserializes_and_maps_to_a_kind ... ok
test msg::tests::inbox_only_shows_messages_addressed_to_me ... ok
test msg::tests::merge_settings_adds_env_and_hook_to_empty ... ok
test msg::tests::merge_settings_block_flag_and_validation ... ok
test msg::tests::history_filters_by_branch_with_thread_closure ... ok
test msg::tests::merge_settings_preserves_existing_and_is_idempotent ... ok
test msg::tests::history_filters_by_participant_and_limit ... ok
test msg::tests::last_view_resolves_numbers_for_the_right_agent ... ok
test msg::tests::late_arriving_older_message_is_still_delivered ... ok
test msg::tests::legacy_watermark_cursor_migrates_to_seen_ids ... ok
test msg::tests::peek_does_not_advance_cursor ... ok
test msg::tests::new_sends_carry_version_and_inferred_kind ... ok
test msg::tests::read_state_is_isolated_per_agent_in_one_clone ... ok
test msg::tests::peek_then_mark_seen_equals_advance ... ok
test msg::tests::remove_branch_scoped_message_noop_when_none_match ... ok
test msg::tests::remove_branch_scoped_preserves_the_roster ... ok
test msg::tests::sanitize_display_neutralises_csi_and_osc_sequences ... ok
test msg::tests::sanitize_display_neutralises_escapes_and_newlines ... ok
test msg::tests::sanitize_display_neutralises_overwrite_and_control_chars ... ok
test msg::tests::remove_branch_scoped_drops_only_that_branch_messages ... ok
test msg::tests::resolve_identity_rejects_poisoned_stored_value ... ok
test msg::tests::roster_records_participants ... ok
test msg::tests::reply_to_untagged_thread_stays_untagged ... ok
test msg::tests::scoped_merge_none_when_branch_empty_and_no_base ... ok
test msg::tests::reply_inherits_thread_branch_not_responder_checkout ... ok
test msg::tests::scoped_merge_keeps_only_branch_messages ... ok
test msg::tests::scoped_merge_preserves_base_messages_and_roster ... ok
test msg::tests::send_auto_tags_current_branch ... ok
test msg::tests::send_then_inbox_delivers_and_advances_cursor ... ok
test msg::tests::stats_report_total_and_tip ... ok
test msg::tests::tag_is_persisted_and_round_trips ... ok
test msg::tests::union_merge_deduplicates_and_orders ... ok
test msg::tests::threads_for_branch_empty_when_no_match ... ok
test msg::tests::unborn_head_leaves_branch_untagged ... ok
test msg::tests::union_merge_is_associative ... ok
test msg::tests::union_merge_is_commutative ... ok
test msg::tests::union_merge_is_idempotent ... ok
test msg::tests::validate_name_enforces_i5h_charset ... ok
test msg::tests::v1_message_round_trips_through_jsonl ... ok
test msg::tests::threads_for_branch_caps_and_reports_total ... ok
test msg::tests::threads_for_branch_filters_and_includes_full_thread ... ok
test objects::tests::blob_path_is_sharded ... ok
test msg::tests::write_agent_seen_unions_with_disk ... ok
test msg::tests::write_identity_rejects_invalid_names ... ok
test msg::tests::union_merge_commits_reconciles_divergent_tips ... ok
test objects::tests::binary_payload_is_stored_with_marker_summary ... ok
test mcp::tests::capture_run_tool_runs_and_returns_summary ... ok
test objects::tests::clamp_highlights_caps_count_and_length ... ok
test objects::tests::clamp_text_truncates_on_char_boundary ... ok
test objects::tests::env_id_matches_full_agentslug_and_bare_slug ... ok
test objects::tests::extract_paths_finds_file_line_refs ... ok
test mcp::tests::capture_run_tool_stores_small_failures ... ok
test objects::tests::gc_dry_run_evicts_nothing ... ok
test objects::tests::absent_blob_degrades_gracefully ... ok
test objects::tests::git_ref_store_get_rejects_corrupt_entry ... ok
test objects::tests::git_ref_store_put_rejects_mismatched_digest ... ok
test objects::tests::capture_records_branch_and_mentioned_files ... ok
test objects::tests::git_ref_store_round_trips ... ok
test objects::tests::capture_without_parser_falls_back_to_generic_structured ... ok
test objects::tests::merge_prefers_valid_incoming_over_corrupt_local ... ok
test objects::tests::capture_populates_structured_result ... ok
test objects::tests::capture_stores_raw_and_restores_exact_bytes ... ok
test objects::tests::parse_duration_units ... ok
test objects::tests::gc_removes_orphans_but_keeps_referenced ... ok
test objects::tests::pull_install_sanitizes_corrupt_only_incoming ... ok
test objects::tests::fsck_flags_absent_and_orphans ... ok
test objects::tests::read_through_command_detection ... ok
test objects::tests::put_repairs_a_corrupt_existing_entry ... ok
test objects::tests::mirror_local_to_gitref_then_recover_after_local_eviction ... ok
test objects::tests::remove_branch_scoped_advances_ref_to_empty_when_all_match ... ok
test objects::tests::mirror_local_heals_a_corrupt_gitref_entry ... ok
test objects::tests::remove_branch_scoped_also_drops_env_evidence_by_env_id ... ok
test objects::tests::remove_branch_scoped_is_noop_when_nothing_matches ... ok
test objects::tests::remove_branch_scoped_drops_only_that_branch ... ok
test objects::tests::resolve_manifest_errors_on_ambiguous_prefix ... ok
test objects::tests::scoped_merge_keeps_only_branch_manifests_when_no_base ... ok
test objects::tests::scoped_merge_includes_env_evidence_by_env_id ... ok
test objects::tests::scoped_merge_none_when_branch_empty_and_no_base ... ok
test objects::tests::search_and_semantics_apply_per_finding ... ok
test objects::tests::search_combines_manifest_and_finding_level_filters ... ok
test objects::tests::search_empty_query_matches_all_findings ... ok
test objects::tests::search_falls_back_to_summary_when_no_structured ... ok
test objects::tests::search_filters_by_env_id ... ok
test objects::tests::search_filters_severity_and_kind ... ok
test objects::tests::search_finding_constraint_with_no_match_excludes_capture ... ok
test objects::tests::search_fingerprint_matches_by_prefix ... ok
test objects::tests::search_fingerprint_prefix_is_case_sensitive ... ok
test objects::tests::search_manifest_level_branch_status_tool ... ok
test objects::tests::search_matches_paths_in_the_locations_vec ... ok
test objects::tests::search_no_criteria_returns_every_capture ... ok
test objects::tests::search_path_matches_by_suffix_either_direction ... ok
test objects::tests::search_preserves_input_order ... ok
test objects::tests::search_query_matches_detail_expected_actual_and_id ... ok
test objects::tests::search_query_matches_message_rule_and_path ... ok
test objects::tests::search_query_only_in_summary_when_structured_findings_dont_match ... ok
test objects::tests::search_returns_every_matching_finding_in_a_capture ... ok
test objects::tests::search_rule_is_exact_case_insensitive ... ok
test objects::tests::search_since_boundary_is_inclusive ... ok
test objects::tests::search_since_is_lexical_cutoff ... ok
test objects::tests::search_status_and_tool_exclude_captures_without_structure ... ok
test objects::tests::sha256_is_stable ... ok
test objects::tests::scoped_merge_preserves_other_branches_already_on_base ... ok
test objects::tests::union_merge_drops_corrupt_incoming_entries ... ok
test policy::tests::commit_policy_no_provenance ... ok
test policy::tests::commit_policy_passes_with_meta ... ok
test policy::tests::glob_double_star ... ok
test policy::tests::glob_exact_match ... ok
test policy::tests::glob_star_in_segment ... ok
test policy::tests::min_message_len_violation ... ok
test policy::tests::path_policy_require_provenance ... ok
test pr::tests::badges_omit_optional_segments_when_zero ... ok
test pr::tests::badges_pluralize_and_format_thousands ... ok
test pr::tests::clean_milestones_caps_at_take_and_walks_newest_first ... ok
test pr::tests::clean_milestones_drops_autotrace_and_session_end ... ok
test pr::tests::clean_milestones_drops_single_file_mechanical_entries ... ok
test pr::tests::clean_milestones_handles_unchecked_boxes_and_blanks ... ok
test pr::tests::collect_duplicate_rows_only_picks_duplicated_code_triggers ... ok
test pr::tests::collect_secret_rows_only_picks_credential_leak_triggers ... ok
test pr::tests::dag_stats_count_files_kinds_and_ratios ... ok
test pr::tests::dag_stats_handle_zero_divisions ... ok
test pr::tests::dag_stats_inline_inverts_ratio_when_edits_dominate ... ok
test pr::tests::dag_stats_inline_omits_empty_signals ... ok
test pr::tests::detective_hero_falls_back_to_observe_when_no_think ... ok
test pr::tests::detective_hero_has_dedicated_stats_section ... ok
test pr::tests::detective_hero_lays_out_four_acts ... ok
test pr::tests::duplicate_pass_note_is_quieter_than_security_pass ... ok
test pr::tests::duplicate_section_uses_warning_alert_and_groups_files ... ok
test pr::tests::empty_sections_render_to_empty_string ... ok
test pr::tests::estimate_commit_cost_prefers_usage_model_over_outer_name ... ok
test pr::tests::estimate_commit_cost_uses_input_and_output_rates ... ok
test pr::tests::excerpt_redacts_token_split_by_control_bytes ... ok
test pr::tests::excerpt_uses_first_nonempty_line_and_redacts ... ok
test pr::tests::extract_file_pulls_path_from_observe_act ... ok
test pr::tests::extract_file_returns_none_for_non_file_traces ... ok
test pr::tests::format_cost_scales_resolution_with_magnitude ... ok
test pr::tests::format_tokens_thousands_breakpoint ... ok
test objects::tests::pinned_referenced_blob_survives_ttl_gc ... ok
test pr::tests::fyi_internal_chatter_hidden_by_default ... ok
test pr::tests::md_escape_neutralizes_html_and_table_injection ... ok
test pr::tests::message_line_shows_review_excerpt ... ok
test pr::tests::minimal_hero_omits_marketing_flourish ... ok
test pr::tests::malformed_ts_is_escaped ... ok
test pr::tests::model_price_picks_tier_by_family_substring ... ok
test pr::tests::parses_duplicate_detail ... ok
test pr::tests::msg_bodies_flag_reveals_fyi_body ... ok
test pr::tests::parses_secret_detail_with_path ... ok
test pr::tests::parses_secret_detail_without_path ... ok
test pr::tests::prompt_maturity_section_renders_headline_and_breakdown ... ok
test pr::tests::receipt_hero_headlines_and_card ... ok
test pr::tests::receipt_hero_omits_blank_signals_gracefully ... ok
test pr::tests::receipt_hero_stat_card_carries_all_signals ... ok
test pr::tests::rejects_duplicate_detail_garbage ... ok
test pr::tests::rejects_secret_detail_garbage ... ok
test pr::tests::replay_hero_emits_fallback_when_dag_empty ... ok
test pr::tests::prompt_maturity_low_coverage_warns ... ok
test pr::tests::replay_hero_emits_goal_label_and_stats ... ok
test pr::tests::review_focus_prioritizes_duplicate_files_before_hot_files ... ok
test pr::tests::review_hero_leads_with_reviewer_triage ... ok
test pr::tests::review_reasoning_highlights_prefer_think_then_note ... ok
test pr::tests::review_typed_kinds_get_excerpt_fyi_does_not ... ok
test objects::tests::non_command_capture_has_no_structured ... ok
test pr::tests::secret_banner_pluralizes_for_single_leak ... ok
test pr::tests::replay_hero_renders_goal_then_dag_then_trail ... ok
test pr::tests::secret_banner_screams_block_merge ... ok
test pr::tests::secret_pass_callout_uses_tip_alert_and_shouts_about_security ... ok
test pr::tests::secret_section_pluralizes_correctly ... ok
test pr::tests::secret_section_uses_caution_alert_and_table ... ok
test pr::tests::short_ts_formats_rfc3339 ... ok
test pr::tests::swimlane_empty_dag_renders_nothing ... ok
test pr::tests::swimlane_expanded_drops_details_wrapper ... ok
test pr::tests::swimlane_labels_shorten_absolute_paths ... ok
test pr::tests::swimlane_mermaid_emits_subgraphs_and_lane_arrows ... ok
test pr::tests::swimlane_sanitizes_dangerous_chars_in_labels_and_titles ... ok
test pr::tests::swimlanes_bucket_by_file_with_reasoning_first ... ok
test pr::tests::swimlanes_compress_consecutive_same_kind_in_file_lane ... ok
test pr::tests::swimlanes_compress_keeps_distinct_kinds_separate ... ok
test pr::tests::swimlanes_overflow_into_other_lane ... ok
test pr::tests::token_section_omits_when_no_captures_or_no_saving ... ok
test pr::tests::token_section_escapes_tool_names_in_table ... ok
test pr::tests::token_section_reports_branch_savings ... ok
test pr::tests::truncate_chars_appends_ellipsis_only_when_cut ... ok
test prompt_score::tests::abstains_on_no_authored_request ... ok
test pr::tests::unknown_kind_is_escaped ... ok
test pr::tests::unknown_kind_strips_control_bytes ... ok
test prompt_score::tests::abstains_on_non_english ... ok
test prompt_score::tests::branch_coverage_low_confidence ... ok
test prompt_score::tests::branch_empty_when_no_prompts ... ok
test prompt_score::tests::branch_drops_unscored_prompts ... ok
test prompt_score::tests::branch_aggregation_is_length_weighted_and_shrunk ... ok
test prompt_score::tests::clarity_band_penalizes_extremes ... ok
test prompt_score::tests::code_detection_and_masking ... ok
test prompt_score::tests::core_is_multiplicative_with_floor ... ok
test prompt_score::tests::diag_calibration ... ignored
test prompt_score::tests::branch_shrinkage_pulls_small_samples_toward_prior ... ok
test prompt_score::tests::empty_and_flags ... ok
test prompt_score::tests::english_short_prompt_is_scored_not_abstained ... ok
test prompt_score::tests::branch_weighted_mean_unaffected_by_precompute ... ok
test prompt_score::tests::enrichment_weights_sum_to_one ... ok
test prompt_score::tests::evidence_bonus_is_additive_and_framing_tops_it_up ... ok
test prompt_score::tests::examples_slot_detected ... ok
test prompt_score::tests::fenced_interior_is_artifact_but_fence_structure_survives ... ok
test prompt_score::tests::executable_acceptance_beats_bare_keyword ... ok
test prompt_score::tests::levels_partition_score_range ... ok
test prompt_score::tests::lexicon_hits_counts_repeats_and_multiword ... ok
test prompt_score::tests::keyword_stuffing_is_capped ... ok
test prompt_score::tests::lexicon_hits_gate_excludes_code_only_words ... ok
test prompt_score::tests::mattr_and_mtld_reward_variety ... ok
test prompt_score::tests::missing_core_slot_caps_the_score ... ok
test prompt_score::tests::mostly_paste_flag_fires_on_thin_ask_around_log_wall ... ok
test prompt_score::tests::prohibitions_without_objective_flag_and_cap ... ok
test prompt_score::tests::readability_numbers_populated ... ok
test prompt_score::tests::repetition_penalty_engages ... ok
test prompt_score::tests::rich_prompt_beats_vague_by_wide_margin ... ok
test prompt_score::tests::scoring_is_deterministic_and_order_independent ... ok
test prompt_score::tests::segments_python_traceback_java_frames_and_diffs ... ok
test prompt_score::tests::segments_rust_panic_from_prose ... ok
test prompt_score::tests::long_prompt_does_not_panic_and_scores_in_range ... ok
test prompt_score::tests::structured_prose_is_not_misread_as_artifact ... ok
test prompt_score::tests::structured_multistep_scores_structure ... ok
test prompt_score::tests::syllable_and_polysyllable_single_pass_is_correct ... ok
test prompt_score::tests::syllables_basic ... ok
test prompt_score::tests::vague_prompt_is_nascent ... ok
test prompt_score::tests::tactical_prompt_survives_missing_context ... ok
test prompt_score::tests::weak_words_lower_objective_and_grounding ... ok
test radio::tests::agent_colors_are_stable ... ok
test prompt_score::tests::word_counts_matches_naive_split ... ok
test radio::tests::empty_log_shows_awaiting_banner ... ok
test radio::tests::frame_is_width_safe_even_with_styling_and_long_bodies ... ok
test radio::tests::proof_uses_global_stats_not_scoped_view_or_message_id ... ok
test radio::tests::rendered_frame_has_no_newlines ... ok
test radio::tests::roster_marks_self_and_derives_activity ... ok
test radio::tests::tiny_terminal_degrades_without_panicking ... ok
test radio::tests::view_ids_track_the_visible_tail_oldest_first ... ok
test prompt_score::tests::mattr_fast_matches_naive ... ok
test recap::tests::parse_keeps_only_away_summary_records ... ok
test recap::tests::split_extracts_goal_and_next ... ok
test recap::tests::split_without_next_returns_full_body_as_summary ... ok
test recap::tests::dry_run_does_not_persist ... ok
test recap::tests::import_is_idempotent ... ok
test repository::tests::copy_scoped_notes_is_nondestructive_over_seeded_dest ... ok
test repository::tests::copy_scoped_notes_only_copies_reachable_commits ... ok
test repository::tests::copy_scoped_notes_zero_when_no_local_notes_ref ... ok
test repository::tests::h5i_log_since_scopes_to_base_exclusive ... ok
test repository::tests::record_human_prompt_accumulates_across_turns ... ok
test repository::tests::test_blame_line_mode ... ok
test repository::tests::remove_notes_deletes_only_listed_commits ... ok
test repository::tests::remove_notes_is_zero_without_notes_ref ... ok
test objects::tests::capture_clamps_oversized_summary_and_highlights ... ok
test repository::tests::test_get_content_at_oid ... ok
test repository::tests::test_load_h5i_record_fallback_to_git ... ok
test repository::tests::test_commit_with_ai_metadata ... ok
test repository::tests::test_scan_test_metrics_detection ... ok
test repository::tests::test_repository_open_initializes_directories ... ok
test repository::tests::test_merge_file_three_way_clean ... ok
test resume::tests::build_suggested_prompt_fallback_when_empty ... ok
test resume::tests::build_suggested_prompt_includes_goal ... ok
test resume::tests::build_suggested_prompt_includes_next_milestone ... ok
test resume::tests::build_suggested_prompt_mentions_risky_file ... ok
test resume::tests::build_suggested_prompt_multiple_done_milestones ... ok
test resume::tests::compute_risky_files_skips_empty_file ... ok
test resume::tests::compute_risky_files_ranks_by_risk_score ... ok
test resume::tests::compute_risky_files_top_phrase_is_most_frequent ... ok
test resume::tests::shorten_path_exact_max_unchanged ... ok
test resume::tests::shorten_path_long_path_truncated ... ok
test resume::tests::shorten_path_short_path_unchanged ... ok
test resume::tests::strip_milestone_marker_plain_text_unchanged ... ok
test resume::tests::strip_milestone_marker_removes_bullet_only ... ok
test repository::tests::unique_commits_empty_for_missing_branch ... ok
test resume::tests::strip_milestone_marker_removes_checkbox_done ... ok
test resume::tests::strip_milestone_marker_removes_checkbox_todo ... ok
test resume::tests::word_wrap_empty_string_returns_empty ... ok
test resume::tests::word_wrap_short_text_is_single_line ... ok
test resume::tests::word_wrap_single_long_word_is_not_split ... ok
test resume::tests::word_wrap_wraps_at_boundary ... ok
test risk::tests::clean_env_scores_zero_info ... ok
test risk::tests::detects_docker_sock ... ok
test risk::tests::detects_privilege_tooling ... ok
test risk::tests::detects_sensitive_targets ... ok
test risk::tests::egress_denied_is_critical_blocked ... ok
test risk::tests::egress_denied_through_classify_env_is_critical ... ok
test risk::tests::first_raw_ip_parsing ... ok
test risk::tests::lane_counts_are_reported_per_lane ... ok
test risk::tests::output_denial_markers ... ok
test risk::tests::process_tier_host_net_only_downgrades_net_lane ... ok
test risk::tests::process_tier_net_deny_greys_nothing ... ok
test risk::tests::raw_ip_egress_is_flagged_over_plain_exfil ... ok
test risk::tests::repeated_probe_bonus_across_runs ... ok
test risk::tests::score_caps_at_100 ... ok
test risk::tests::three_lane_spread_escalates_to_critical ... ok
test risk::tests::violation_event_drives_critical_with_denial_ts ... ok
test risk::tests::workspace_tier_downgrades_pressure_to_weak_isolation ... ok
test rules::tests::ci_cd_silent_with_ci_intent ... ok
test rules::tests::ci_cd_fires_without_intent ... ok
test rules::tests::code_execution_fires_on_eval ... ok
test rules::tests::code_execution_fires_on_os_system ... ok
test rules::tests::code_execution_skips_comment_lines ... ok
test rules::tests::config_file_detected ... ok
test rules::tests::config_file_excludes_lockfile ... ok
test rules::tests::credential_leak_ignores_placeholder_without_value ... ok
test rules::tests::credential_leak_ignores_short_lines ... ok
test rules::tests::duplicated_code_fires_on_repeated_block ... ok
test rules::tests::credential_leak_fires_on_pem_header ... ok
test rules::tests::duplicated_code_ignores_comments_and_whitespace ... ok
test rules::tests::duplicated_code_ignores_short_block ... ok
test rules::tests::duplicated_code_skips_allowlisted_path ... ok
test rules::tests::credential_leak_fires_on_api_key_assignment ... ok
test rules::tests::is_test_path_recognises_layouts_we_claim ... ok
test rules::tests::duplicated_code_skips_test_files ... ok
test rules::tests::duplicated_code_still_fires_on_non_test_paths ... ok
test rules::tests::large_diff_silent_under_threshold ... ok
test rules::tests::large_diff_fires_over_threshold ... ok
test rules::tests::lockfile_fires_without_dep_intent ... ok
test rules::tests::lockfile_silent_with_dep_intent ... ok
test rules::tests::permission_change_fires_on_chmod_777 ... ok
test rules::tests::refactor_anomaly_fires_on_large_net_addition ... ok
test rules::tests::permission_change_skips_comment ... ok
test rules::tests::refactor_anomaly_silent_without_refactor_intent ... ok
test rules::tests::scope_expansion_fires_when_extra_files_changed ... ok
test rules::tests::scope_expansion_silent_when_no_file_mentioned ... ok
test rules::tests::undeclared_deletion_fires_on_heavy_deletion ... ok
test rules::tests::undeclared_deletion_silent_when_refactor_stated ... ok
test sandbox::tests::agent_profile_injects_is_sandbox ... ok
test sandbox::tests::agent_codex_profile_scopes_to_codex_only ... ok
test sandbox::tests::agent_profile_refuses_tiers_that_cannot_enforce_egress ... ok
test sandbox::tests::agent_runtime_from_identity_maps_codex_else_claude ... ok
test sandbox::tests::builtin_passes_term_for_interactive_sessions ... ok
test sandbox::tests::builtin_agent_profile_loads_without_policy_file ... ok
test sandbox::tests::config_lock_paths_picks_existing_project_dirs_and_home_files ... ok
test repository::tests::unique_commits_excludes_commits_shared_with_other_branches ... ok
test sandbox::tests::effective_auto_honors_a_declared_tier_without_probing ... ok
test sandbox::tests::empty_private_paths_keeps_policy_digest_stable ... ok
test sandbox::tests::egress_allowlist_under_process_fails_closed ... ok
test sandbox::tests::fs_deny_lint_allows_disjoint_grants ... ok
test sandbox::tests::fs_deny_lint_rejects_granted_parent_of_denied_child ... ok
test sandbox::tests::fsize_changes_the_policy_digest ... ok
test sandbox::tests::effective_auto_skips_container_without_an_image ... ok
test sandbox::tests::injected_is_sandbox_is_not_overridden ... ok
test sandbox::tests::empty_tools_allowlist_allows_anything ... ok
test sandbox::tests::isolation_claim_parse_and_order ... ok
test sandbox::tests::isolation_override_wins_over_profile ... ok
test sandbox::tests::missing_named_profile_is_an_error ... ok
test sandbox::tests::net_egress_allowed_under_container_refused_under_process ... ok
test sandbox::tests::non_agent_profile_does_not_inject_is_sandbox ... ok
test sandbox::tests::missing_policy_file_yields_builtin_workspace_default ... ok
test sandbox::tests::parse_mem_units ... ok
test sandbox::tests::parse_wall_units ... ok
test sandbox::tests::parses_the_design_doc_example_profile ... ok
test sandbox::tests::policy_digest_is_stable_and_content_sensitive ... ok
test sandbox::tests::persona_sources_parse_and_validate ... ok
test sandbox::tests::private_paths_parse_with_kind_and_persist_defaults ... ok
test sandbox::tests::profile_declared_isolation_reads_the_field ... ok
test sandbox::tests::private_paths_reject_unsafe_and_overlapping ... ok
test sandbox::tests::resolve_container_requires_runtime_and_image ... ok
test sandbox::tests::resolve_process_refused_off_linux ... ok
test sandbox::tests::resolve_process_requires_landlock_and_seccomp ... ok
test sandbox::tests::resolve_refuses_unimplemented_backends ... ok
test sandbox::tests::resolve_workspace_needs_nothing ... ok
test sandbox::tests::resources_fsize_and_cpu_parse_and_default_off ... ok
test sandbox::tests::seccomp_deny_list_covers_security_critical_syscalls ... ok
test sandbox::tests::seccomp_deny_program_builds ... ok
test sandbox::tests::secret_grant_bad_source_fails_closed ... ok
test sandbox::tests::secret_grants_are_accepted_and_normalized ... ok
test sandbox::tests::shell_rcfile_parses_and_defaults_none ... ok
test sandbox::tests::supervised_builtin_is_confined_and_net_deny ... ok
test sandbox::tests::tools_allowlist_enforced_when_non_empty ... ok
test sandbox::tests::user_defined_agent_profile_merges_over_agent_builtin ... ok
test objects::tests::redact_flag_scrubs_payload_and_command_before_storage ... ok
test sandbox::tests::run_records_resource_usage ... ok
test sandbox_policy::tests::enum_toml_spellings_are_strict ... ok
test sandbox_policy::tests::process_profile_defaults_are_fail_closed ... ok
test sandbox_policy::tests::resolved_policy_round_trips_through_toml ... ok
test sandbox_policy::tests::serde_defaults_keep_missing_optional_policy_fields_conservative ... ok
test seccomp_notify::tests::abi_struct_sizes_are_the_stable_layout ... ok
test seccomp_notify::tests::bpf_program_shape ... ok
test seccomp_notify::tests::ioctl_numbers_are_well_formed ... ok
test objects::tests::redact_flag_off_leaves_payload_intact ... ok
test sandbox::tests::workspace_run_executes_in_workdir_with_wall_clock ... ok
test seccomp_notify::tests::live_serve_with_pidfd_self_terminates ... ok
test secrets::tests::anthropic_key_fires ... ok
test seccomp_notify::tests::pidfd_signals_child_exit ... ok
test secrets::tests::connection_string_in_test_file_path_skipped ... ok
test secrets::tests::aws_access_key_id_fires ... ok
test secrets::tests::github_pat_fires ... ok
test secrets::tests::http_basic_auth_url_does_not_fire_on_plain_url ... ok
test secrets::tests::high_entropy_assignment_fires ... ok
test secrets::tests::http_basic_auth_url_fires ... ok
test secrets::tests::jdbc_password_param_fires ... ok
test secrets::tests::jwt_fires ... ok
test secrets::tests::keywords_match_any_of_many ... ok
test secrets::tests::keyword_prefilter_blocks_unanchored_lines ... ok
test secrets::tests::keywords_match_empty_means_always_run ... ok
test secrets::tests::keywords_match_with_keyword_absent ... ok
test secrets::tests::keywords_match_with_keyword_present ... ok
test secrets::tests::low_entropy_assignment_suppressed ... ok
test secrets::tests::mongodb_connection_string_fires ... ok
test secrets::tests::path_allowlist_skips_image ... ok
test secrets::tests::path_allowlist_skips_lockfile ... ok
test secrets::tests::placeholder_suppressed ... ok
test secrets::tests::mysql_connection_string_fires ... ok
test secrets::tests::postgres_connection_string_fires ... ok
test secrets::tests::postgres_connection_string_long_form_scheme ... ok
test secrets::tests::private_key_pem_fires ... ok
test secrets::tests::postgres_rejects_low_entropy_password ... ok
test secrets::tests::redact_text_leaves_clean_text_untouched ... ok
test secrets::tests::redact_text_fails_closed_on_stoplist_word ... ok
test secrets::tests::redact_text_preserves_line_structure ... ok
test secrets::tests::redact_text_scrubs_multiple_same_rule_secrets ... ok
test secrets::tests::redis_connection_string_fires_with_empty_user ... ok
test secrets::tests::shannon_entropy_basics ... ok
test secrets::tests::vendor_dir_skipped ... ok
test secrets::tests::redact_text_scrubs_secret_keeps_prose ... ok
test secrets_broker::tests::command_extractor_caps_output_fail_closed ... ok
test secrets_broker::tests::command_extractor_refused_unless_allowed ... ok
test secrets_broker::tests::command_extractor_nonzero_exit_fails_closed ... ok
test seccomp_notify::tests::live_socket_gate_denies_raw_allows_inet ... ok
test secrets_broker::tests::command_extractor_runs_when_allowed ... ok
test secrets_broker::tests::default_source_is_namespaced_env_var ... ok
test secrets_broker::tests::empty_value_fails_closed ... ok
test secrets_broker::tests::env_inject_brokers_value_and_records_no_value ... ok
test secrets_broker::tests::file_inject_refused_off_workspace_tier ... ok
test secrets_broker::tests::command_extractor_under_limits_ok ... ok
test secrets_broker::tests::file_inject_writes_0600_and_points_env_at_it ... ok
test secrets_broker::tests::fingerprint_is_stable_and_value_free ... ok
test secrets_broker::tests::missing_source_fails_closed ... ok
test secrets_broker::tests::file_source_trims_trailing_newline ... ok
test secrets_broker::tests::multiple_grants_all_brokered_independently ... ok
test secrets_broker::tests::one_missing_grant_fails_the_whole_broker ... ok
test secrets_broker::tests::resolves_env_source ... ok
test session_log::tests::test_analyze_session_bash_commands_captured ... ok
test session_log::tests::test_analyze_session_churn_score ... ok
test session_log::tests::test_analyze_session_consulted_file_count ... ok
test session_log::tests::test_analyze_session_footprint_edited_files ... ok
test session_log::tests::test_analyze_session_detects_uncertainty_phrase ... ok
test session_log::tests::test_analyze_session_implicit_deps ... ok
test session_log::tests::test_analyze_session_key_decisions_extracted ... ok
test session_log::tests::test_analyze_session_message_count ... ok
test session_log::tests::test_analyze_session_no_uncertainty_in_confident_text ... ok
test session_log::tests::test_analyze_session_replay_hash_differs_for_different_content ... ok
test session_log::tests::test_analyze_session_replay_hash_is_stable ... ok
test session_log::tests::test_analyze_session_tool_call_count ... ok
test session_log::tests::test_dedup_empty_input ... ok
test session_log::tests::test_analyze_session_user_trigger ... ok
test session_log::tests::test_dedup_keeps_distinct_items ... ok
test session_log::tests::test_dedup_removes_exact_duplicates ... ok
test session_log::tests::test_extract_snippet_contains_phrase ... ok
test session_log::tests::test_extract_snippet_phrase_at_start ... ok
test session_log::tests::test_extract_snippet_respects_max_len ... ok
test session_log::tests::test_floor_char_boundary_inside_multibyte ... ok
test session_log::tests::test_group_confidence_values_collected ... ok
test session_log::tests::test_group_empty_input ... ok
test session_log::tests::test_group_riskiest_file_first ... ok
test session_log::tests::test_group_single_annotation ... ok
test session_log::tests::test_jaccard_completely_disjoint ... ok
test session_log::tests::test_jaccard_empty_strings ... ok
test session_log::tests::test_jaccard_identical_strings ... ok
test session_log::tests::test_jaccard_partial_overlap ... ok
test session_log::tests::test_pointer_string_empty ... ok
test session_log::tests::test_pointer_string_overlap_skips_second ... ok
test session_log::tests::test_pointer_string_single ... ok
test session_log::tests::test_pointer_string_two_non_overlapping ... ok
test session_log::tests::test_shorten_path_exact_length ... ok
test session_log::tests::test_extract_snippet_truncates_on_char_boundary ... ok
test session_log::tests::test_shorten_path_fits ... ok
test session_log::tests::test_shorten_path_truncates_with_ellipsis ... ok
test session_log::tests::test_shorten_path_with_multibyte ... ok
test session_log::tests::test_split_sentences_short_fragments_dropped ... ok
test session_log::tests::test_split_sentences_three_sentences ... ok
test session_log::tests::test_split_sentences_trailing_fragment_kept_if_long_enough ... ok
test session_log::tests::test_timeline_empty_annotations ... ok
test session_log::tests::test_timeline_middle_annotation ... ok
test session_log::tests::test_timeline_riskiest_wins_on_collision ... ok
test session_log::tests::test_timeline_single_annotation_lands_at_start ... ok
test session_log::tests::test_timeline_two_annotations_at_endpoints ... ok
test session_log::tests::test_timeline_zero_width ... ok
test storage::tests::doctor_repair_creates_layout_and_schema_version ... ok
test storage::tests::doctor_export_writes_manifest_and_sidecar_copy ... ok
test storage::tests::doctor_reports_missing_sidecar_without_repair ... ok
test storage::tests::store_io_error_is_actionable_on_permission_denied ... ok
test storage::tests::ensure_layout_actionable_when_parent_unwritable ... ok
test storage::tests::store_owner_hint_reports_uid_mismatch ... ok
test structured::tests::cap_limits_findings_and_records_total ... ok
test storage::tests::unwritable_objects_dir_yields_actionable_put_error ... ok
test structured::tests::cargo_unknown_subcommand_declines ... ok
test structured::tests::compact_render_is_one_line_per_finding_and_smaller ... ok
test structured::tests::compact_render_shows_truncation_and_generic_body ... ok
test structured::tests::detail_cap_respects_byte_limit_and_records_omission ... ok
test structured::tests::cargo_test_parser_extracts_panic_with_assertion ... ok
test structured::tests::eslint_parser_extracts_problems_with_file ... ok
test structured::tests::extra_carries_non_diagnostic_tool_data ... ok
test structured::tests::fingerprint_is_stable_across_line_shifts ... ok
test structured::tests::generic_success_is_ok_not_passed ... ok
test structured::tests::diagnostic_parsers_decline_on_garbage ... ok
test structured::tests::go_build_failure_surfaces_compiler_diagnostic ... ok
test structured::tests::go_bare_package_fail_declines_to_generic ... ok
test structured::tests::mypy_parser_extracts_errors_with_codes ... ok
test structured::tests::pytest_all_pass_is_passed_status ... ok
test structured::tests::pytest_no_tests_ran_is_handled_not_declined ... ok
test structured::tests::pytest_parser_declines_without_summary_line ... ok
test structured::tests::go_test_parser_extracts_failures ... ok
test structured::tests::pytest_parser_extracts_findings_and_status ... ok
test structured::tests::ruff_clean_is_ok ... ok
test structured::tests::status_never_passes_on_nonzero_exit ... ok
test structured::tests::ruff_parser_extracts_diagnostics_and_fixable ... ok
test structured::tests::yaml_quotes_risky_scalars ... ok
test structured::tests::tsc_parser_extracts_diagnostics ... ok
test supervisor::tests::nft_empty_allowlist_drops_everything_but_lo ... ok
test supervisor::tests::nft_rule_pins_the_exact_port ... ok
test structured::tests::yaml_render_round_trips_through_json ... ok
test supervisor::tests::nft_ruleset_is_default_drop_with_allowlist ... ok
test supervisor::tests::pin_egress_parses_host_and_explicit_port ... ok
test supervisor::tests::resolve_egress_pins_hostnames_not_ip_literals ... ok
test supervisor::tests::probe_is_fail_closed_and_explained ... ok
test supervisor::tests::socket_gate_is_default_deny ... ok
test supervisor::tests::socketpair_gate_allows_anonymous_unix_pairs ... ok
test supervisor::tests::run_egress_fails_closed_when_unsupported ... ok
test team::tests::add_env_writes_team_identity_for_inbox_routing ... ok
test team::tests::apply_agent_errors_for_unknown_or_unsubmitted_agent ... ok
test team::tests::apply_agent_applies_latest_submission_without_finalize ... ok
test team::tests::auto_create_assigns_generated_persona_keys_not_the_runtime ... ok
test team::tests::auto_create_roster_derives_per_team_env_slugs ... ok
test team::tests::auto_peer_review_excludes_non_submitter_with_allow_missing ... ok
test objects::tests::identical_content_dedupes ... ok
test team::tests::auto_peer_review_freezes_grants_and_instructs ... ok
test team::tests::auto_peer_review_rejects_bad_artifact_kind_before_freezing ... ok
test team::tests::auto_peer_review_ingests_staged_submission_before_freezing ... ok
test team::tests::compare_tolerates_env_absent_locally ... ok
test team::tests::auto_peer_review_requires_two_submitters ... ok
test sandbox::tests::effective_auto_never_picks_an_unrunnable_tier ... ok
test team::tests::current_team_pointer_roundtrips ... ok
test team::tests::create_add_submit_freeze_projects_from_events ... ok
test objects::tests::ttl_gc_evicts_referenced_but_stale ... ok
test team::tests::discuss_refused_before_freeze ... ok
test sandbox::tests::home_bind_shadows_target_inside_confined_child ... ok
test team::tests::discussion_marks_later_submission_as_influenced ... ok
test secrets_broker::tests::command_extractor_times_out_fail_closed ... ok
test team::tests::dispatch_to_unbound_agent_is_a_noop_not_an_error ... ok
test team::tests::gen_agent_id_avoids_collisions ... ok
test team::tests::dispatch_does_not_block_add_env_submit_or_freeze ... ok
test team::tests::review_instruction_is_read_only_and_drops_sealed_commands ... ok
test team::tests::dispatch_grant_and_review_are_recorded ... ok
test team::tests::freeze_refuses_missing_submission ... ok
test team::tests::find_submission_resolves_artifact_and_diffs_against_base ... ok
test team::tests::latest_submission_is_newest_by_time_not_lexicographic_id ... ok
test team::tests::finalize_refuses_divergent_verifier_commands ... ok
test team::tests::submit_auto_snapshots_dirty_worktree ... ok
test team::tests::submit_refuses_no_op_submission ... ok
test team::tests::submit_review_in_open_round_records_but_does_not_deliver ... ok
test team::tests::submit_review_post_freeze_delivers_to_target_inbox_and_marks_influence ... ok
test team::tests::sync_outbound_ingests_staged_submission_live ... ok
test mcp::tests::env_lifecycle_over_mcp ... FAILED
test sandbox::tests::wall_clock_kill_fires ... ok
test token_filter::tests::classify_detects_diff ... ok
test token_filter::tests::classify_detects_json ... ok
test token_filter::tests::classify_detects_test ... ok
test team::tests::apply_winner_replays_submission_diff_and_records_target_commit ... ok
test token_filter::tests::detect_tool_handles_wrappers_and_subcommands ... ok
test token_filter::tests::cargo_all_pass_shrinks_but_not_empty ... ok
test token_filter::tests::cargo_keeps_error_blocks ... ok
test token_filter::tests::cargo_test_failure_keeps_panic_and_is_not_labeled_ok ... ok
test token_filter::tests::line_score_ranks_errors_above_noise ... ok
test token_filter::tests::normalize_template_collapses_numbers ... ok
test token_filter::tests::cargo_metadata_json_falls_back_to_generic ... ok
test team::tests::verify_and_finalize_selects_passing_candidate ... ok
test token_filter::tests::pytest_summary_line_required_else_none ... ok
test team::tests::worker_once_finalizes_verifier_ready_run ... ok
test team::tests::verify_applies_revised_multi_commit_submission ... ok
test token_filter::tests::strip_ansi_collapses_bare_cr ... ok
test token_filter::tests::strip_ansi_removes_color_codes ... ok
test token_filter::tests::dedup_collapses_identical_runs ... ok
test token_filter::tests::git_diff_adds_file_count_header ... ok
test token_filter::tests::json_summary_surfaces_error_fields ... ok
test token_filter::tests::pytest_all_pass_shrinks_to_one_line ... ok
test token_filter::tests::small_output_is_returned_intact ... ok
test token_filter::tests::scored_filter_keeps_errors_and_elides_noise ... ok
test token_filter::tests::pytest_keeps_buried_failures ... ok
test token_filter::tests::template_folding_collapses_parameterized_log_lines ... ok
test token_filter::tests::filter_routes_through_adapter_when_cmd_set ... ok
test token_filter::tests::token_budget_trims_middle ... ok
test container::tests::shim_detects_combined_c_flags_and_skips_non_commands ... ok

failures:

---- mcp::tests::env_lifecycle_over_mcp stdout ----

thread 'mcp::tests::env_lifecycle_over_mcp' (89818) panicked at src/mcp.rs:2802:14:
called `Result::unwrap()` on an `Err` value: Metadata error: supervised spawn failed: Invalid argument (os error 22)
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace


failures:
    mcp::tests::env_lifecycle_over_mcp

test result: FAILED. 939 passed; 1 failed; 1 ignored; 0 measured; 0 filtered out; finished in 37.83s beats a
bare 'ensure'). Direction weights seeded from arXiv 2601.13118 prevalence.

Negative smells (#4): objective folds a vagueness penalty and reads ~0 for
prohibitions-only prompts (new NoObjective flag + coach line); constraints
split into must-class (weighted up) vs should-class; ambiguous 'otherwise'
conditionals dock direction.

Abstention (#5): score_prompt returns unscored(reason) for no-authored-request
(pure paste) or non-English text (function-word ratio) rather than a confident
wrong number; rendered as 'unscored (reason)' in the PR body and API.

Shrinkage (#7): score_branch shrinks the length-weighted mean toward a neutral
prior (empirical Bayes) so one or two prompts can't crown or condemn a branch;
unscored prompts drop from the mean but still count against coverage.

Recalibrated to keep vague/prohibition/stuffed prompts nascent-developing and
concrete+bounded asks proficient. 42 module + 5 integration tests green;
clippy clean on default and --no-default-features.