fix(cli): include loopback tools in cli prompts by clawsweeper[bot] · Pull Request #83828 · openclaw/openclaw

clawsweeper · 2026-05-19T00:40:58Z

Makes #83785 merge-ready for the ClawSweeper automerge loop.
The edit pass should inspect the live PR diff, review comments, and failing checks; rebase if needed; keep the contributor branch credited; and stop only when validation is green or an external blocker is proven.
Known failing checks:

Failing check: Security High (network-ssrf-boundary):FAILURE (https://github.com/openclaw/openclaw/actions/runs/26064671153/job/76632683943)
Failing check: dependency-change-awareness:FAILURE (https://github.com/openclaw/openclaw/actions/runs/26064670128/job/76632680812)

ClawSweeper 🐠 replacement reef notes:

Cluster: automerge-openclaw-openclaw-83785
Source PRs: fix(cli): include loopback tools in cli prompts #83785
Credit: Source PR: fix(cli): include loopback tools in cli prompts #83785
Validation: pnpm check:changed
Replacement reason: ClawSweeper could not update the source PR branch directly, so it opened a writable replacement PR instead.
Automerge requested by: @Takhoffman

Repair fallback: GitHub rejected the repair branch push because it updates workflow files and the ClawSweeper app token does not have workflows permission

Inherited issue-closing references from the source PR:
Fixes #83210

Co-author credit kept:

@TurboTheTurtle: Co-authored-by: Andy Ye 35905412+TurboTheTurtle@users.noreply.github.com

fish notes: model gpt-5.5, reasoning high; reviewed against d196564.

clawsweeper · 2026-05-19T00:42:00Z

Codex review: passed.

Workflow note: Future ClawSweeper reviews update this same comment in place.

How this review workflow works

ClawSweeper keeps one durable marker-backed review comment per issue or PR.
Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
Maintainers can also comment @clawsweeper review to request a fresh review only.
Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

Summary
The PR feeds loopback-scoped MCP tools into CLI system prompts and reports, persists a prompt tool-name hash for CLI session reuse, adds regression tests, and adds a changelog entry.

Reproducibility: yes. from source inspection: current main builds the CLI prompt and report with tools: [], while memory prompt guidance is gated on memory_search or memory_get being present in availableTools. I did not execute a live CLI turn in this read-only review, but the source path and source PR terminal proof line up.

PR rating
Overall: 🐚 platinum hermit
Proof: 🦞 diamond lobster
Patch quality: 🐚 platinum hermit
Summary: Strong terminal proof and focused tests support a normal-good patch rating, with the overall tier capped by the intentional session-state upgrade risk.

Rank-up moves:

none

What the crustacean ranks mean

🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

PR egg
✨ Hatched: 🥚 common Mossy Patch Peep

       /\  .---.  /\         
      /  \/     \/  \        
     /   ( -   - )   \       
    |       ._.       |      
    |   /|  ===  |\   |      
     \  \|______/|/  /       
      '._  `--'  _.'         
         '-.__.-'            
       _/|_|  |_|\_          
      /__|      |__\         
       .-----------.         
      '-------------'

Rarity: 🥚 common.
Trait: finds missing screenshots.
Share on X: post this hatch
Copy: My PR egg hatched a 🥚 common Mossy Patch Peep in ClawSweeper.

What is this egg doing here?

Eggs appear after the PR passes real-behavior proof. It is here for vibes, not verdicts: it does not change labels, ratings, merge decisions, or automation.
The shell reacts to review momentum: open follow-up work warms it up, re-review makes it wobble, and a clean final review lets it hatch.
How to hatch it: reach status: 👀 ready for maintainer look or status: 🚀 automerge armed; that usually means sufficient real-behavior proof, no blocking P0/P1/P2 findings, no security attention needed, and clean correctness.
The hatch is seeded from this repository and PR number, so the same PR keeps the same creature; the reviewed head SHA can only change safe visual details.
Rarity is just collectible sparkle: 🥚 common, 🌱 uncommon, 💎 rare, ✨ glimmer, and 🌈 legendary.

Real behavior proof
Sufficient (terminal): The source PR includes terminal output from an after-fix live loopback prompt proof showing nonempty resolved/report tool surfaces and prompt_has_empty_tool_surface=false.

Risk before merge
Why this matters: - Merging intentionally invalidates provider-side CLI session reuse when the loopback prompt tool-name surface changes, so existing CLI-backed sessions may start a fresh provider conversation once after upgrade.

Maintainer options:

Accept the prompt-surface reset (recommended)
Merge through the normal exact-head gates while accepting that CLI-backed sessions may refresh once when their prompt-visible loopback tools change.
Run a resumed CLI smoke first
If maintainers want extra upgrade confidence, run one resumed bundled-MCP CLI turn and confirm the prompt/report tool surface is nonempty and the session reset is expected.

Next step before merge
No repair lane is needed because there are no actionable code-review findings; the remaining action is normal merge/automerge gating plus maintainer acceptance of the session reset risk.

Security
Cleared: The diff does not change workflows, dependencies, package resolution, secrets handling, or downloaded code execution; the new helper preserves existing loopback tool policy and owner-only filtering.

Review details

Best possible solution:

Land the resolver-reuse approach with maintainer acceptance of the one-time CLI session refresh, then let the linked issue close through the PR's closing reference after merge.

Do we have a high-confidence way to reproduce the issue?

Yes from source inspection: current main builds the CLI prompt and report with tools: [], while memory prompt guidance is gated on memory_search or memory_get being present in availableTools. I did not execute a live CLI turn in this read-only review, but the source path and source PR terminal proof line up.

Is this the best way to solve the issue?

Yes: reusing the same loopback-scoped resolver that serves MCP tools is the narrowest maintainable fix and avoids a second tool-policy implementation. The session hash addition is appropriate for prompt correctness, with the session reset called out as an upgrade risk rather than a patch defect.

Label justifications:

P2: This fixes a normal-priority CLI agent prompt/session bug with limited blast radius and focused tests.
merge-risk: 🚨 session-state: The PR deliberately changes CLI session reuse fingerprints, which can refresh existing provider-side CLI sessions on upgrade.

What I checked:

Current-main bug path: Current main still passes tools: [] into buildCliAgentSystemPrompt, so prompt sections gated on available tools see an empty set. (src/agents/cli-runner/prepare.ts:365, 2c8f78e723ec)
Prompt gating contract: The memory prompt section returns no guidance unless availableTools contains memory_search or memory_get, matching the linked issue's root cause. (extensions/memory-core/src/prompt-section.ts:7, 2c8f78e723ec)
Patch implementation path: The PR resolves loopback-scoped tools only when bundled MCP has an active loopback runtime, hashes the sorted tool names, passes the tools into prompt construction, and uses the hash in CLI session reuse decisions. (src/agents/cli-runner/prepare.ts:287, d196564d4db2)
Loopback resolver reuse: The PR extracts the existing loopback tool-resolution policy, preserving the loopback surface, native tool exclusions, and owner-only policy before both prompt use and MCP cache use. (src/gateway/mcp-http.runtime.ts:22, d196564d4db2)
Regression coverage: The added tests cover the positive loopback prompt-tools path, system prompt report entries, prompt tool-name hash invalidation, and the unavailable-runtime fallback that must not advertise tools. (src/agents/cli-runner/prepare.test.ts:993, d196564d4db2)
Real behavior proof: The source PR provided terminal proof from an after-fix live loopback prompt run: 24 resolved/report tools and prompt_has_empty_tool_surface=false. (2d4b16aa7168)

Likely related people:

steipete: Git history shows the largest share of work on the central CLI runner, CLI session, and MCP loopback files, including the CLI backend fallback, Claude CLI loopback bridge, MCP loopback split, and CLI skills prompt wiring. (role: primary recent area contributor; confidence: high; commits: d8f1124d594f, 3de09fbe7427, 25b069a6f3cd; files: src/agents/cli-runner/prepare.ts, src/agents/cli-session.ts, src/gateway/mcp-http.runtime.ts)
Takhoffman: The recent CLI session invalidation commit is directly adjacent to this PR's prompt-hash session reuse behavior, and the PR body records Takhoffman as the automerge requester. (role: recent session-state contributor; confidence: medium; commits: f8edd09a2c08; files: src/agents/cli-session.ts, src/agents/cli-runner/prepare.ts)
vincentkoc: History on the same central files includes recent auth/runtime hardening, cycle extraction, and cli-runner test seam work, making this a useful secondary routing candidate for agent/runtime boundary review. (role: adjacent runtime contributor; confidence: medium; commits: 859eb0666282, 74e7b8d47b18, 7008379ff085; files: src/agents/cli-runner/prepare.ts, src/agents/cli-session.ts, src/gateway/mcp-http.runtime.ts)

Codex review notes: model gpt-5.5, reasoning high; reviewed against 2c8f78e723ec.

clawsweeper · 2026-05-19T00:49:22Z

🦞✅
ClawSweeper merged this PR after the passing review.

Source: clawsweeper[bot]
Feedback: structured ClawSweeper verdict: pass (sha=d196564d4db2cec15d16b5c9ace8d258848dac15)
Merge status: merged by ClawSweeper automerge
Merged at: 2026-05-19T01:30:07Z
Merge commit: 48d9966aa1de

What merged:

The PR feeds loopback-scoped MCP tools into CLI system prompts and reports, persists a prompt tool-name hash for CLI session reuse, adds regression tests, and adds a changelog entry.
Reproducibility: yes. from source inspection: current main builds the CLI prompt and report with tools: [] ... execute a live CLI turn in this read-only review, but the source path and source PR terminal proof line up.

Automerge notes:

PR branch already contained follow-up commit before automerge: fix(cli): gate prompt loopback tools on active runtime
PR branch already contained follow-up commit before automerge: fix(cli): include loopback tools in cli prompts

The automerge loop is complete.

Automerge progress:

2026-05-19 01:29:52 UTC review passed d196564d4db2 (structured ClawSweeper verdict: pass (sha=d196564d4db2cec15d16b5c9ace8d258848da...)

2026-05-19 00:49:22 UTC repair queued d196564d4db2 (autonomous) Run: https://github.com/openclaw/clawsweeper/actions/runs/26069285607

2026-05-19 00:53:14 UTC repair started (running) in 0s Run: https://github.com/openclaw/clawsweeper/actions/runs/26069285607 automerge-openclaw-openclaw-83785

2026-05-19 00:53:19 UTC repair completed d196564d4db2 (no branch change) in 5s Run: https://github.com/openclaw/clawsweeper/actions/runs/26069285607 no planned fix actions

2026-05-19 00:53:18 UTC merge check queued d196564d4db2 (checks and exact-head review are ready)

2026-05-19 01:22:59 UTC review queued d196564d4db2 (queued)

2026-05-19 01:30:09 UTC merged d196564d4db2 (merged by ClawSweeper automerge)

Summary: - The PR feeds loopback-scoped MCP tools into CLI system prompts and reports, persists a prompt tool-name hash for CLI session reuse, adds regression tests, and adds a changelog entry. - Reproducibility: yes. from source inspection: current main builds the CLI prompt and report with `tools: []` ... execute a live CLI turn in this read-only review, but the source path and source PR terminal proof line up. Automerge notes: - PR branch already contained follow-up commit before automerge: fix(cli): gate prompt loopback tools on active runtime - PR branch already contained follow-up commit before automerge: fix(cli): include loopback tools in cli prompts Validation: - ClawSweeper review passed for head d196564d4db2cec15d16b5c9ace8d258848dac15. - Required merge gates passed before the squash merge. Prepared head SHA: d196564d4db2cec15d16b5c9ace8d258848dac15 Review: openclaw/openclaw#83828 (comment) Co-authored-by: Andy Ye <35905412+TurboTheTurtle@users.noreply.github.com> Co-authored-by: clawsweeper <274271284+clawsweeper[bot]@users.noreply.github.com> Co-authored-by: clawsweeper[bot] <274271284+clawsweeper[bot]@users.noreply.github.com> Approved-by: takhoffman Co-authored-by: takhoffman <781889+takhoffman@users.noreply.github.com>

Summary: - The PR feeds loopback-scoped MCP tools into CLI system prompts and reports, persists a prompt tool-name hash for CLI session reuse, adds regression tests, and adds a changelog entry. - Reproducibility: yes. from source inspection: current main builds the CLI prompt and report with `tools: []` ... execute a live CLI turn in this read-only review, but the source path and source PR terminal proof line up. Automerge notes: - PR branch already contained follow-up commit before automerge: fix(cli): gate prompt loopback tools on active runtime - PR branch already contained follow-up commit before automerge: fix(cli): include loopback tools in cli prompts Validation: - ClawSweeper review passed for head d196564. - Required merge gates passed before the squash merge. Prepared head SHA: d196564 Review: openclaw#83828 (comment) Co-authored-by: Andy Ye <35905412+TurboTheTurtle@users.noreply.github.com> Co-authored-by: clawsweeper <274271284+clawsweeper[bot]@users.noreply.github.com> Co-authored-by: clawsweeper[bot] <274271284+clawsweeper[bot]@users.noreply.github.com> Approved-by: takhoffman Co-authored-by: takhoffman <781889+takhoffman@users.noreply.github.com>

TurboTheTurtle and others added 3 commits May 19, 2026 00:40

fix(cli): include loopback tools in cli prompts

65ff297

fix(cli): gate prompt loopback tools on active runtime

0c81a9d

fix(cli): include loopback tools in cli prompts

d196564

clawsweeper Bot mentioned this pull request May 19, 2026

fix(cli): include loopback tools in cli prompts #83785

Closed

openclaw-barnacle Bot removed the proof: supplied External PR includes structured after-fix real behavior proof. label May 19, 2026

clawsweeper Bot added status: 🚀 automerge armed This PR is in ClawSweeper's automerge lane. and removed status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. labels May 19, 2026

clawsweeper Bot merged commit 48d9966 into main May 19, 2026
185 of 204 checks passed

clawsweeper Bot deleted the clawsweeper/automerge-openclaw-openclaw-83785 branch May 19, 2026 01:30

Haderach-Ram mentioned this pull request May 19, 2026

Ecosystem Digest — 2026-05-19 Haderach-Ram/openclaw-radar#12

Open

github-actions Bot mentioned this pull request May 19, 2026

📡 Upstream Digest — 2026-05-19 02:33 UTC curtismercier/openclaw-mods#895

Open

clawsweeper Bot mentioned this pull request May 19, 2026

Populate codex-cli tool entries in system prompt report #65409

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix(cli): include loopback tools in cli prompts#83828

fix(cli): include loopback tools in cli prompts#83828
clawsweeper[bot] merged 3 commits into
mainfrom
clawsweeper/automerge-openclaw-openclaw-83785

clawsweeper Bot commented May 19, 2026

Uh oh!

clawsweeper Bot commented May 19, 2026 •

edited

Loading

Uh oh!

clawsweeper Bot commented May 19, 2026 •

edited

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Uh oh!

Conversation

clawsweeper Bot commented May 19, 2026

Uh oh!

clawsweeper Bot commented May 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clawsweeper Bot commented May 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

clawsweeper Bot commented May 19, 2026 •

edited

Loading

clawsweeper Bot commented May 19, 2026 •

edited

Loading