Discord: resolve /think autocomplete from session model by MonkeyLeeT · Pull Request #49176 · openclaw/openclaw

MonkeyLeeT · 2026-03-17T17:57:30Z

Summary

Describe the problem and fix in 2–5 bullets:

Problem: Discord native /think autocomplete used config-default model context, not the effective Discord route's session model.
Why it matters: users on an xhigh-capable session model (for example openai-codex/gpt-5.4) could miss xhigh in slash choices.
What changed: native command option autocomplete now resolves the effective Discord route, reads provider/model from that route's session override, and passes it into resolveCommandArgChoices.
What did NOT change (scope boundary): no changes to /think command semantics, backend thinking capability logic, or non-Discord channels.

Change Type (select all)

Scope (select all touched areas)

Linked Issue/PR

User-visible / Behavior Changes

Discord native /think autocomplete now reflects the active effective-route session model override rather than only config defaults.

Security Impact (required)

New permissions/capabilities? (No)
Secrets/tokens handling changed? (No)
New/changed network calls? (No)
Command/tool execution surface changed? (No)
Data access scope changed? (No)
If any Yes, explain risk + mitigation:

Repro + Verification

Environment

OS: macOS
Runtime/container: local dev
Model/provider: anthropic default config + openai-codex/gpt-5.4 session override
Integration/channel (if any): Discord native slash commands
Relevant config (redacted): agents.defaults.model.primary=anthropic/claude-sonnet-4.5

Steps

Configure a default model that does not support xhigh.
Set the effective Discord route session override to openai-codex/gpt-5.4.
Trigger Discord native /think autocomplete.

Expected

xhigh appears in autocomplete choices for the effective-route session.

Actual

Before fix: choices reflected default model context and could omit xhigh.
After fix: choices resolve from effective-route session context and include xhigh.

Evidence

Attach at least one:

Failing test/log before + passing after
Trace/log snippets
Screenshot/recording
Perf numbers (if relevant)

Human Verification (required)

What you personally verified (not just CI), and how:

Verified scenarios: targeted Discord native-command test suite passes; added regression test covering session-override-aware /think autocomplete.
Edge cases checked: fallback path when route/session resolution fails (returns null context and preserves prior behavior).
What you did not verify: live Discord runtime interaction against a real guild.

Review Conversations

I replied to or resolved every bot review conversation I addressed in this PR.
I left unresolved only the conversations that still need reviewer or maintainer judgment.

If a bot review conversation is addressed by this PR, resolve that conversation yourself. Do not leave bot review conversation cleanup for maintainers.

Compatibility / Migration

Backward compatible? (Yes)
Config/env changes? (No)
Migration needed? (No)
If yes, exact upgrade steps:

Failure Recovery (if this breaks)

How to disable/revert this change quickly: revert this PR commit.
Files/config to restore: extensions/discord/src/monitor/native-command.ts, extensions/discord/src/monitor/native-command-ui.ts
Known bad symptoms reviewers should watch for: Discord /think autocomplete errors or missing choices.

Risks and Mitigations

List only real risks for this PR. Add/remove entries as needed. If none, write None.

Risk: route/session override resolution can fail in autocomplete context.
- Mitigation: code catches failures and falls back to prior default-model behavior.

greptile-apps · 2026-03-17T18:00:27Z

Greptile Summary

This PR fixes a bug where Discord native /think autocomplete choices were resolved from the config-default model context instead of the bound session's effective model override. The fix introduces resolveDiscordNativeChoiceContext, which resolves the session route and reads the stored model override before calling resolveCommandArgChoices, so providers/models only available on the override (e.g. xhigh on openai-codex/gpt-5.4) appear correctly in autocomplete.

Key changes:

resolveDiscordNativeChoiceContext — new async helper that resolves provider/model from the bound session override, with a try/catch that returns null on failure to preserve prior default-model behaviour.
buildDiscordCommandOptions — now accepts an optional resolveChoiceContext callback; invokes it only for function-based arg.choices, so static choice arrays are unaffected.
resolveDiscordModelPickerRoute — union type extended to include AutocompleteInteraction so it can be reused in the new context resolver.
New test file native-command.think-autocomplete.test.ts verifies the session-override-aware happy path.

Notable concern: the shouldAutocomplete flag is still determined at command-build time using a default-model-context evaluation of resolveCommandArgChoices. If the default model returns zero choices for a function-based arg (not the case for /think today, but possible in future commands), shouldAutocomplete will be false and the session-aware autocomplete handler will never be registered, silently undermining the intent of this fix.

Confidence Score: 4/5

Safe to merge — the change is well-scoped, has a safety fallback, and the primary happy-path is tested.
The fix is logically correct for the described scenario and the fallback (catch → null) prevents regressions. The one structural concern (shouldAutocomplete evaluated without session context) is a latent edge-case risk rather than a bug in the current PR's scope. Test coverage is adequate for the happy path but missing the error/fallback branch.
native-command.ts lines 171–175 (shouldAutocomplete guard using default-model context)

Comments Outside Diff (1)

extensions/discord/src/monitor/native-command.ts, line 171-175 (link)

shouldAutocomplete evaluated with default-model context

The resolvedChoices at line 171 are computed once at command-build time using the config-default model (no session override). The shouldAutocomplete flag therefore depends on the default model returning at least one choice for the arg.

If a session override targets a model that supports a command option that the default model does not support at all (i.e. the default-context call returns an empty array), shouldAutocomplete will be false and the session-aware autocomplete handler introduced by this PR will never be registered — meaning the fix won't fire for that arg.

For the described bug (default model omits xhigh but still returns other thinking levels), this works correctly today. But the same pattern in a scenario where the default model returns zero choices (e.g. a command option that is only meaningful for a specific provider) would silently skip the session-aware path.

Consider guarding with arg.preferAutocomplete === true || typeof arg.choices === "function" as an additional condition so that function-based choice args always enable autocomplete regardless of what the default-model evaluation returns:

const shouldAutocomplete =
  arg.preferAutocomplete === true ||
  typeof arg.choices === "function" ||
  (resolvedChoices.length > 0 && resolvedChoices.length > 25);

Prompt To Fix With AI

This is a comment left during a code review.
Path: extensions/discord/src/monitor/native-command.ts
Line: 171-175

Comment:
**`shouldAutocomplete` evaluated with default-model context**

The `resolvedChoices` at line 171 are computed once at command-build time using the config-default model (no session override). The `shouldAutocomplete` flag therefore depends on the default model returning at least one choice for the arg.

If a session override targets a model that supports a command option that the default model does not support at all (i.e. the default-context call returns an empty array), `shouldAutocomplete` will be `false` and the session-aware autocomplete handler introduced by this PR will never be registered — meaning the fix won't fire for that arg.

For the described bug (default model omits `xhigh` but still returns other thinking levels), this works correctly today. But the same pattern in a scenario where the default model returns *zero* choices (e.g. a command option that is only meaningful for a specific provider) would silently skip the session-aware path.

Consider guarding with `arg.preferAutocomplete === true || typeof arg.choices === "function"` as an additional condition so that function-based choice args always enable autocomplete regardless of what the default-model evaluation returns:

```
const shouldAutocomplete =
  arg.preferAutocomplete === true ||
  typeof arg.choices === "function" ||
  (resolvedChoices.length > 0 && resolvedChoices.length > 25);
```

How can I resolve this? If you propose a fix, please make it concise.

Prompt To Fix All With AI

This is a comment left during a code review.
Path: extensions/discord/src/monitor/native-command.ts
Line: 171-175

Comment:
**`shouldAutocomplete` evaluated with default-model context**

The `resolvedChoices` at line 171 are computed once at command-build time using the config-default model (no session override). The `shouldAutocomplete` flag therefore depends on the default model returning at least one choice for the arg.

If a session override targets a model that supports a command option that the default model does not support at all (i.e. the default-context call returns an empty array), `shouldAutocomplete` will be `false` and the session-aware autocomplete handler introduced by this PR will never be registered — meaning the fix won't fire for that arg.

For the described bug (default model omits `xhigh` but still returns other thinking levels), this works correctly today. But the same pattern in a scenario where the default model returns *zero* choices (e.g. a command option that is only meaningful for a specific provider) would silently skip the session-aware path.

Consider guarding with `arg.preferAutocomplete === true || typeof arg.choices === "function"` as an additional condition so that function-based choice args always enable autocomplete regardless of what the default-model evaluation returns:

```
const shouldAutocomplete =
  arg.preferAutocomplete === true ||
  typeof arg.choices === "function" ||
  (resolvedChoices.length > 0 && resolvedChoices.length > 25);
```

How can I resolve this? If you propose a fix, please make it concise.

---

This is a comment left during a code review.
Path: extensions/discord/src/monitor/native-command.think-autocomplete.test.ts
Line: 28-42

Comment:
**No test coverage for the fallback path**

The PR description notes that "fallback path when route/session resolution fails (returns null context and preserves prior behavior)" was verified, but there is no automated test exercising that path.

Consider adding a second `it` block that sets `mocks.resolveBoundConversationRoute.mockRejectedValue(new Error("network"))` (or `mockResolvedValue(null)`) and asserts that `respond` is still called with non-empty choices (the default-model set). This prevents regressions on the safety catch introduced at line 528 of `native-command.ts`.

How can I resolve this? If you propose a fix, please make it concise.

_{Last reviewed commit: 6a10d30}

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 6a10d30e33

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 3bf90acd01

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-03-17T18:33:00Z

+    const storePath = resolveStorePath(params.cfg.session?.store, {
+      agentId: route.agentId,
+    });
+    const sessionStore = loadSessionStore(storePath);


Avoid cloning full session store on every autocomplete keystroke

This new autocomplete context path calls loadSessionStore for each /think autocomplete event, and even cache hits in loadSessionStore return structuredClone(cached.store) of the entire session map (src/config/sessions/store-cache.ts), so typing in Discord can repeatedly clone a large store on the event loop. In deployments with large sessions.json files or many concurrent users, this can add enough synchronous CPU latency to miss or delay autocomplete responses; this path should avoid whole-store reads (for example by memoizing per interaction window or adding a targeted session-entry lookup).

Useful? React with 👍 / 👎.

Thanks for the review. Agreed on the performance concern category, with one important distinction:

This PR removes the new regression in /think autocomplete: we no longer force loadSessionStore(..., { skipCache: true }) on each event. The autocomplete path now uses loadSessionStore(storePath) (extensions/discord/src/monitor/native-command.ts:517), so it goes through the shared session-store cache path (src/config/sessions/store.ts:200-210).

You’re correct that cache hits still return structuredClone(cached.store) (src/config/sessions/store-cache.ts:60), so there is still synchronous per-call work proportional to store size.

So the acute “forced disk read/parse per keystroke” issue is fixed here; the broader “whole-store clone on cached reads” cost remains a valid follow-up optimization for session-store access generally (including autocomplete). We can track that separately with a targeted API improvement (for example a keyed session-entry read/memoized accessor) rather than block this regression fix.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 6df099c54a

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: e1079c1f8a

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: b8651a8bd5

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: d31aace422

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a17162e239

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 5306e0fb3f

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 4d845cc12a

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-03-23T15:19:18Z

+            typeof arg.choices === "function" && resolveChoiceContext
+              ? await resolveChoiceContext(interaction)
+              : null;


Enforce auth before loading autocomplete session context

The autocomplete handler now calls resolveChoiceContext for function-based choices before any command authorization runs, so /think suggestions are computed from the effective session model even for users who would be denied later by commands.allowFrom, guild/channel restrictions, or DM policy checks in dispatchDiscordCommandInteraction. In restricted servers this leaks bound-session capability metadata (for example whether xhigh appears) to unauthorized users; the context lookup should be gated by the same auth checks or return no sensitive choices when unauthorized.

Useful? React with 👍 / 👎.

Keeping this PR scoped to the original bug fix.

This PR changes /think autocomplete to use the effective-route session model so choices match the command target model. I’m not taking on the broader metadata-hardening change in this PR; that would require sharing the full command auth decision with autocomplete and expands the patch materially.

If we want to harden autocomplete visibility, I’d prefer to do that as a follow-up.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 504e14f952

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-03-24T03:17:39Z

+          const context =
+            typeof arg.choices === "function" && resolveChoiceContext
+              ? await resolveChoiceContext(interaction)


Apply command auth before resolving autocomplete context

The autocomplete handler now calls resolveChoiceContext (which reads the effective route/session model) before any authorization checks run, so users who are later denied in dispatchDiscordCommandInteraction can still probe /think and infer bound-session capability metadata (for example whether xhigh appears). This is a behavior regression from static choices because the sensitive part is now session-derived; gate the context lookup behind the same DM/group/guild allowlist checks used for execution.

Useful? React with 👍 / 👎.

Same call here: keeping this PR scoped to fixing the session-context mismatch for /think autocomplete.

The auth-gating concern is real but broader than this bug fix, and the minimal correct version still adds noticeable complexity. I’d rather defer that hardening to a follow-up than bundle it into this PR.

chatgpt-codex-connector · 2026-03-24T03:17:39Z

+    const storePath = resolveStorePath(params.cfg.session?.store, {
+      agentId: route.agentId,
+    });
+    const sessionStore = loadSessionStore(storePath);


Avoid full session-store loads per autocomplete keystroke

resolveDiscordNativeChoiceContext loads the session store on every autocomplete request, and loadSessionStore returns a cloned full map even on cache hits (readSessionStoreCache does structuredClone). In large sessions.json deployments, repeated /think keystrokes can cause synchronous clone/parse work on the event loop and delay or miss Discord autocomplete responses; this path needs a targeted lookup or short-lived memoization instead of whole-store loads per event.

Useful? React with 👍 / 👎.

Agree this is a broader performance concern category, but I’m keeping this PR limited to correcting the model-context source for /think autocomplete.

This branch already removed the earlier uncached read path; what remains is the general session-store read/cloning behavior, which is not specific to this bug. I’d rather track any targeted lookup/memoization work separately.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a0d66316a3

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

chatgpt-codex-connector · 2026-03-25T01:02:36Z

+  threadBindings: ThreadBindingManager;
+}) {
+  const resolved = await resolveDiscordModelPickerRouteState(params);
+  return resolved.route;


Return effective route for model-picker context

resolveDiscordModelPickerRoute now returns resolved.route, which is the unbound base route from resolveDiscordNativeInteractionRouteState. In bound-thread or configured-binding setups, this drops the session/agent override that the picker UI depends on, so /model picker views can load providers/current model from the wrong session while command execution still targets the effective bound session. This creates inconsistent picker state (including false mismatch warnings) and can surface the wrong model set for the interaction context.

Useful? React with 👍 / 👎.

* Discord: use session model for /think autocomplete * Discord: use cached session store in think autocomplete * Discord: align think autocomplete with effective bound route * Discord: fix think autocomplete route-resolution test mocks * Discord: stabilize think autocomplete CI coverage * Discord: gate think autocomplete context behind auth * Discord: share slash auth for think autocomplete * Discord: localize think autocomplete auth gate * Discord: drop think autocomplete auth gating * Discord: align native autocomplete auth and route readiness * Discord: use effective route for model picker --------- Co-authored-by: Tak Hoffman <[email protected]>

…nto add-json-schema-to-cli * 'add-json-schema-to-cli' of github.com:kvokka/openclaw: (58 commits) test: make vitest config tests platform-aware fix: refresh available tools when the session model changes (openclaw#54184) fix: prefer freshest transcript session owners test: disable Vitest fs cache on Windows fix: prefer deterministic transcript session keys fix: prefer deterministic session usage targets fix: prefer deterministic session id resume targets feat: add /tools runtime availability view (openclaw#54088) fix: enforce sandbox visibility for session_status ids fix: drop spawned visibility list caps fix: verify exact spawned session visibility fix: enforce spawned session visibility in key resolve fix: resolve exact session ids without fuzzy limits Discord: resolve /think autocomplete from session model (openclaw#49176) chore(agents): normalize pi embedded runner imports feat(gateway): make openai compatibility agent-first test(parallel): force unit-fast batch planning test(browser): stabilize default browser detection mocks fix: prefer current parents in session rows fix: prefer current subagent owners in session rows ...

* Discord: use session model for /think autocomplete * Discord: use cached session store in think autocomplete * Discord: align think autocomplete with effective bound route * Discord: fix think autocomplete route-resolution test mocks * Discord: stabilize think autocomplete CI coverage * Discord: gate think autocomplete context behind auth * Discord: share slash auth for think autocomplete * Discord: localize think autocomplete auth gate * Discord: drop think autocomplete auth gating * Discord: align native autocomplete auth and route readiness * Discord: use effective route for model picker --------- Co-authored-by: Tak Hoffman <[email protected]>

greptile-apps Bot reviewed Mar 17, 2026

View reviewed changes

Comment thread extensions/discord/src/monitor/native-command.think-autocomplete.test.ts

openclaw-barnacle Bot added channel: discord Channel integration: discord size: S labels Mar 17, 2026

chatgpt-codex-connector Bot reviewed Mar 17, 2026

View reviewed changes

Comment thread extensions/discord/src/monitor/native-command.ts Outdated

MonkeyLeeT force-pushed the codex/discord-think-session-model branch from 6a10d30 to 3bf90ac Compare March 17, 2026 18:26

chatgpt-codex-connector Bot reviewed Mar 17, 2026

View reviewed changes

MonkeyLeeT force-pushed the codex/discord-think-session-model branch from 3bf90ac to 6df099c Compare March 18, 2026 04:54

chatgpt-codex-connector Bot reviewed Mar 18, 2026

View reviewed changes

Comment thread extensions/discord/src/monitor/native-command-ui.ts Outdated

openclaw-barnacle Bot added size: M and removed size: S labels Mar 18, 2026

chatgpt-codex-connector Bot reviewed Mar 18, 2026

View reviewed changes

Comment thread extensions/discord/src/monitor/native-command.think-autocomplete.test.ts Outdated

chatgpt-codex-connector Bot reviewed Mar 20, 2026

View reviewed changes

Comment thread extensions/discord/src/monitor/native-command-ui.ts Outdated

MonkeyLeeT force-pushed the codex/discord-think-session-model branch from b8651a8 to 2ff1434 Compare March 20, 2026 20:40

chatgpt-codex-connector Bot reviewed Mar 22, 2026

View reviewed changes

Comment thread extensions/discord/src/monitor/native-command.ts

openclaw-barnacle Bot added size: L and removed size: M labels Mar 22, 2026

chatgpt-codex-connector Bot reviewed Mar 22, 2026

View reviewed changes

Comment thread extensions/discord/src/monitor/native-command.ts Outdated

chatgpt-codex-connector Bot reviewed Mar 23, 2026

View reviewed changes

Comment thread extensions/discord/src/monitor/native-command-ui.ts Outdated

Comment thread extensions/discord/src/monitor/native-command-ui.ts Outdated

openclaw-barnacle Bot added size: M and removed size: L labels Mar 23, 2026

chatgpt-codex-connector Bot reviewed Mar 23, 2026

View reviewed changes

MonkeyLeeT force-pushed the codex/discord-think-session-model branch from 4d845cc to 504e14f Compare March 24, 2026 03:09

chatgpt-codex-connector Bot reviewed Mar 24, 2026

View reviewed changes

MonkeyLeeT added 6 commits March 24, 2026 19:28

Discord: use session model for /think autocomplete

0a51367

Discord: use cached session store in think autocomplete

4ca8355

Discord: align think autocomplete with effective bound route

c99f7be

Discord: fix think autocomplete route-resolution test mocks

a2577c9

Discord: stabilize think autocomplete CI coverage

339e872

Discord: gate think autocomplete context behind auth

8a3bd74

MonkeyLeeT and others added 4 commits March 24, 2026 19:28

Discord: share slash auth for think autocomplete

1ec8fbe

Discord: localize think autocomplete auth gate

085878a

Discord: drop think autocomplete auth gating

402cb1b

Discord: align native autocomplete auth and route readiness

a0d6631

Takhoffman force-pushed the codex/discord-think-session-model branch from 2b43cea to a0d6631 Compare March 25, 2026 00:58

openclaw-barnacle Bot added size: L and removed size: M labels Mar 25, 2026

chatgpt-codex-connector Bot reviewed Mar 25, 2026

View reviewed changes

Discord: use effective route for model picker

e69aebf

Takhoffman merged commit 5799322 into openclaw:main Mar 25, 2026
9 checks passed

github-actions Bot mentioned this pull request Mar 25, 2026

📡 Upstream Digest — 2026-03-25 01:24 UTC curtismercier/openclaw-mods#355

Open

MonkeyLeeT deleted the codex/discord-think-session-model branch March 31, 2026 20:26

This was referenced Jun 21, 2026

fix(channels): resolve native /think menu levels via runtime catalog for live-discovered models #94067

Merged

fix(discord): resolve redundant type constituents in native-command-route #80473

Closed

Uh oh!

Uh oh!

Conversation

MonkeyLeeT commented Mar 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Change Type (select all)

Scope (select all touched areas)

Linked Issue/PR

User-visible / Behavior Changes

Security Impact (required)

Repro + Verification

Environment

Steps

Expected

Actual

Evidence

Human Verification (required)

Review Conversations

Compatibility / Migration

Failure Recovery (if this breaks)

Risks and Mitigations

Uh oh!

greptile-apps Bot commented Mar 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Summary

Confidence Score: 4/5

Comments Outside Diff (1)

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Mar 17, 2026

Choose a reason for hiding this comment

Uh oh!

MonkeyLeeT Mar 17, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

Uh oh!

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

MonkeyLeeT commented Mar 17, 2026 •

edited

Loading

greptile-apps Bot commented Mar 17, 2026 •

edited

Loading