Skip to content
This repository was archived by the owner on Nov 13, 2026. It is now read-only.

AGENTS.md: frontier route 3 via Paseo's antigravity provider; acp-dispatch is a shim (djbclark-ade slim-down C3) - #6

Merged
djbclark merged 2 commits into
masterfrom
slim-c3-2026-11
Nov 6, 2026
Merged

djbclark merged 2 commits into
masterfrom
slim-c3-2026-11

Conversation

@djbclark

@djbclark djbclark commented Nov 6, 2026 •

Copy link
Copy Markdown
Collaborator

djbclark-ade's slim-down (docs/slim/PLAN.md, C3 per-file migration, J2b) retires the agy TUI on 2026-11-06; the Google AI plan is reached through Paseo's antigravity provider. This PR changes only AGENTS.md (5 insertions, 4 deletions), the same five edits as frdminc/tendcf's:

  • Frontier route 3 (Gemini 4 Argon) is now a paseo run --provider antigravity --model gemini-4-argon --no-wait --title NAME --cwd WORKTREE "$(cat BRIEF)" line (paseo run has no --prompt-file; the brief ends with the report path, read after paseo wait AGENT_ID). The antigravity full-access mode id is to be confirmed before passing --mode.
  • The weekly Carbon recheck uses paseo provider models antigravity instead of acp-run agy --info.
  • "agy's Opus pool" reads "the antigravity provider's Opus pool".
  • A dated sentence says acp-dispatch is a shim over paseo run, to be replaced by paseo run lines in a follow-up; every other acp-dispatch line is unchanged.

Opened by the slim-down lead under the operator's J2b grant (2026-11-06).


Generated by Claude Code


Note

Low Risk
Documentation-only refresh; no runtime behavior or security surface touched.

Overview
Updates the standing-rule section of AGENTS.md so it reflects djbclark-ade’s 2026-11-06 slim-down: it now states acp-dispatch is a temporary paseo run shim and will be swapped for explicit paseo run invocations later while keeping existing quota safeguards.

It also retargets references from “agy” to Paseo’s antigravity provider—Sonnet slices may use the antigravity Opus pool, Gemini Argon reviews spell out the eventual acp-dispatch antigravity … command, insist those reviews stay behind --kind frontier (never bare paseo run), and remind reviewers there is still no read-only Google provider, and the weekly Gemini Carbon recheck now runs via paseo provider models antigravity because the agy TUI/acp-run agy --info path is retired.

Reviewed by Cursor Bugbot for commit 3d34a54. Bugbot is set up for automated code reviews on this repo. Configure here.

…(slim-down C3)

The agy TUI is retired (2026-11-06, E9). Route 3 and the Carbon recheck use
paseo run / paseo provider models; acp-dispatch noted as a shim.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BvGHM3M9bXWXHwGGHz6Ctd
@cursor

cursor Bot commented Nov 6, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: 6e8b848f-d6fb-41a0-bd39-67ccfdb09d39)

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Nov 6, 2026 •

Copy link
Copy Markdown
Codex Review Summary: ✅ Completed

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-11-06T22:44:01.541777Z 3d34a54 Manual request
🔒 Security Review ✅ Completed 2026-11-06T22:43:17.735329Z 3d34a54 Manual request
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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 0c611bbd65

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".

Comment thread AGENTS.md Outdated
1. astra: `acp-dispatch codex --model gpt-6-astra --kind frontier --effort high --perm read-only --frontier-reason "<repo> standing rule: cross-model review" --name NAME -f BRIEF -C WORKTREE`
2. Fable: `acp-dispatch claude --model fable --kind frontier --effort high --perm read-only --frontier-reason "..." --name NAME -f BRIEF -C WORKTREE`
3. Gemini 4 Argon: `acp-dispatch agy --model gemini-4-argon --kind frontier --effort high --perm all --frontier-reason "<repo> standing rule: cross-model review" --name NAME -f BRIEF -C WORKTREE`. acp-run has no verified read-only mode for agy, so put read-only instructions in BRIEF and check the worktree diff before and after the review. That restricts Argon by instruction only, not by a read-only sandbox. Argon is fallback-only: it stands in when astra or Fable is short. Its frontier wiring landed in djbclark-ade PR 224 (merged 2026-11-02).
3. Gemini 4 Argon (a Paseo run since 2026-11-06; the agy TUI is retired, slim-down E9, and the Google AI plan is reached through Paseo's antigravity provider): `paseo run --provider antigravity --model gemini-4-argon --no-wait --title NAME --cwd WORKTREE "$(cat BRIEF)"`. `paseo run` takes the brief as its positional prompt (it has no `--prompt-file`; `paseo send` does), so BRIEF is read in by the shell. End BRIEF with the report path ("write your full report to REPORT_PATH") and read that file after `paseo wait AGENT_ID`. The old `--perm all` is `--mode`, whose ids are provider-specific: pass the antigravity provider's full-access mode only after confirming its id (not verified 2026-11-06); there is no verified read-only mode, so put read-only instructions in BRIEF and check the worktree diff before and after the review. That restricts Argon by instruction only, not by a read-only sandbox. Argon is fallback-only: it stands in when astra or Fable is short. Its frontier wiring landed in djbclark-ade PR 224 (merged 2026-11-02); a `paseo run` bypasses `acp-dispatch`'s preflight, so check the Google pool with `aiuse --available --json` first.

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Preserve guarded dispatch for the Argon fallback

When astra or Fable is short, this replacement directs the reviewer to launch Argon through paseo run instead of the mandatory acp-dispatch --kind frontier path. Checking aiuse --available --json first does not preserve the required estimate-floor, in-flight reservation, and per-wave frontier-cap checks. The new fallback therefore bypasses safeguards that this same document explicitly says remain mandatory. Keep Argon behind guarded dispatch, or provide an equivalent guarded replacement before directing sessions to use it.

AGENTS.md reference: AGENTS.md:L109-L114

Useful? React with 👍 / 👎.

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Confirmed and fixed in 3d34a54. Route 3 stays behind acp-dispatch --kind frontier (estimate floor, in-flight reservation, per-wave cap, exit 5): it reads acp-dispatch agy … today and acp-dispatch antigravity … once the djbclark-ade slim-down lands, whose acp-dispatch shim maps the agent name to the Paseo provider. The bare paseo run line and the aiuse --available --json substitute are gone.


Generated by Claude Code

…ex P1); the agent name becomes the Paseo provider when the slim-down lands
@cursor

cursor Bot commented Nov 6, 2026

Copy link
Copy Markdown

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: 562859ad-c05b-4570-9427-0ee7481a43fe)

djbclark commented Nov 6, 2026

Copy link
Copy Markdown
Collaborator Author

@codex review


Generated by Claude Code

@chatgpt-codex-connector

Copy link
Copy Markdown

🛡️ Codex Security Review · Automatically triggered

Security review completed. No security issues were found in this pull request.

Reviewed commit: 3d34a549b2

View security finding report

Only the user who started this review can view the report in Codex.

ℹ️ About Codex security reviews in GitHub

This is an experimental Codex feature. Security reviews are triggered when:

  • You comment "@codex security review"
  • A regular code review gets triggered (for example, "@codex review" or when a PR is opened), and you’re opted in so security review runs alongside code review

Once complete, Codex will leave suggestions, or a comment if no findings are found.

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. 🚀

Reviewed commit: 3d34a549b2

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".

@djbclark
djbclark merged commit bf2842e into master Nov 6, 2026
2 checks passed
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants