Add native 'h5i team auto-peer-review' command#188
Conversation
Native equivalent of scripts/team-review.sh: in one shot sync staged submissions, freeze the open round, grant every candidate mutual review access, and send each the review-and-revise instruction. Scopes the review circle to agents that actually submitted this round (new TeamAutoReviewReport.reviewers): requires >=2 submitters before the one-way freeze and excludes non-submitters, so --allow-missing runs review among the submitters instead of aborting at grant_review on an agent with no submission. Artifact kinds are validated before freezing.
🪙 100% AI-authored · 4 files · prompt 45/100
Important 🎯 Goal: add native 'h5i team auto-peer-review' command equivalent to scripts/team-review.sh (sync→freeze→mutual grant→review instructions)
💬 The ask
📍 What shipped
Note 🪴 Prompt maturity: 45/100 · developing · 1 prompt scored (100% of AI commits) 📊 heuristic breakdown
Readability (display-only): Flesch 61 · FK grade 9.8 · Gunning Fog 13.5 Offline classical-NLP score — no LLM. Higher = more specific, bounded, and grounded prompting. Tip ✅ Security scan cleanNo credentials leaked in this branch's diff. h5i scanned every added line against the secret rule pack and found nothing to rotate. Warning Duplicate code introduced in 1 file — consider extracting a helper before this lands.
🧠 Reasoning by file — 9683 nodes across 5 lanes, latest 24 only%%{init: {"flowchart": {"nodeSpacing": 42, "rankSpacing": 48, "diagramPadding": 14}, "themeVariables": {"fontSize": "18px"}} }%%
flowchart TB
subgraph lane_0["💭 Reasoning · 4 steps"]
direction LR
n_lane_0_0["THINK · I'll explore the existing team functionality and the referenced script b…"]:::t
n_lane_0_1["THINK · Now the handler. I'll add it before the `TeamCommands::Review` handler a…"]:::t
n_lane_0_2["THINK · Good question to push on — let me critically assess what I actually cove…"]:::t
n_lane_0_3["THINK · On branch `auto-review` (not main), so I'll commit directly."]:::t
n_lane_0_0 --> n_lane_0_1
n_lane_0_1 --> n_lane_0_2
n_lane_0_2 --> n_lane_0_3
end
subgraph lane_1["📄 src/team.rs · 14 ops"]
direction LR
n_lane_1_0["READ × 4"]:::o
n_lane_1_1["EDIT × 3"]:::a
n_lane_1_2["READ"]:::o
n_lane_1_3["EDIT × 6"]:::a
n_lane_1_0 --> n_lane_1_1
n_lane_1_1 --> n_lane_1_2
n_lane_1_2 --> n_lane_1_3
end
subgraph lane_2["📄 src/main.rs · 3 ops"]
direction LR
n_lane_2_0["EDIT × 3"]:::a
end
subgraph lane_3["📄 README.md · 2 ops"]
direction LR
n_lane_3_0["READ"]:::o
n_lane_3_1["EDIT"]:::a
n_lane_3_0 --> n_lane_3_1
end
subgraph lane_4["📄 src/msg.rs · 1 op"]
direction LR
n_lane_4_0["READ"]:::o
end
classDef o fill:#dbeafe,stroke:#1e3a8a,color:#0b1c4a;
classDef t fill:#fef3c7,stroke:#92400e,color:#3f2d05;
classDef a fill:#dcfce7,stroke:#166534,color:#0a2e16;
classDef n fill:#ede9fe,stroke:#5b21b6,color:#221251;
classDef m fill:#e5e7eb,stroke:#374151,color:#0b0f17;
Note 🪙 Token reduction — 1 captured tool output kept out of context: 2132 → 423 tokens (80% saved, 1709 tokens). Full output is recoverable with 📜 Per-commit provenance
|
Native equivalent of scripts/team-review.sh: in one shot sync staged submissions, freeze the open round, grant every candidate mutual review access, and send each the review-and-revise instruction.
Scopes the review circle to agents that actually submitted this round (new TeamAutoReviewReport.reviewers): requires >=2 submitters before the one-way freeze and excludes non-submitters, so --allow-missing runs review among the submitters instead of aborting at grant_review on an agent with no submission. Artifact kinds are validated before freezing.