autopilot

An automated workflow for taking a GitHub issue in the Cloudflare vinext project through fixing, review, and merging. A GitHub issue is a tracked task or bug, and a pull request is the proposed code change for it.

In plain words
What is it for?
Handling a numbered vinext issue from start to merge. It creates a worktree and branch, installs dependencies, makes and tests the fix, opens a pull request, and addresses review feedback.
Why use it?
Moving an issue to a merged change requires many manual steps, from creating a branch and running checks to responding to review feedback. This organizes those steps into one process.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/cloudflare/vinext/autopilot
Clone the repo
git clone --depth 1 https://github.com/cloudflare/vinext
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 750 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00018 $0.00750
Opus 5 $0.00009 $0.00375
Sonnet 5 $0.00004 $0.00150
Haiku 4.5 $0.00002 $0.00075

Measured yesterday against content hash c87b3996c488, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

autopilot scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.opencode/commands/autopilot.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Run the full issue-to-merge workflow for GitHub issue #$ARGUMENTS on cloudflare/vinext.

Phase 1: Fix the issue

  1. Run gh issue view $ARGUMENTS to understand the issue.

  2. Generate a short slug from the issue title (lowercase, hyphens, max 30 chars).

  3. Create a worktree and branch:

    git fetch origin main
    git worktree add ../vinext-fix-$ARGUMENTS -b fix/$ARGUMENTS-<slug> origin/main
    
  4. Install dependencies: run pnpm install in the worktree.

  5. Fix the issue in ../vinext-fix-$ARGUMENTS/. Follow AGENTS.md guidelines.

    • If touching server code, check dev/prod parity across all server files.
  6. Run all checks from the worktree: pnpm run build && pnpm test && pnpm run typecheck && pnpm run lint

  7. Commit: fix: <description> (#$ARGUMENTS)

  8. Push: git push -u origin fix/$ARGUMENTS-<slug>

  9. Create the PR:

    gh pr create --title "fix: <description>" --body "Fixes #$ARGUMENTS
    
    ## Summary
    <what changed and why>
    
    ## Test plan
    <what tests were added/updated>"
    
  10. Note the PR number — you need it for Phase 2.

Phase 2: Independent review

Invoke the reviewer subagent to get an independent review from a different model:

@reviewer Review pull request #<PR_NUMBER> on cloudflare/vinext.

1. Run gh pr view <PR_NUMBER> to understand the PR context.
2. Run gh pr diff <PR_NUMBER> to see all changes.
3. Read the full source files that were modified for surrounding context.
4. Check server parity if any server files were touched.
5. Post your review using gh pr review with inline comments.

Wait for the reviewer to finish before proceeding.

Phase 3: Address review feedback

  1. Read all review comments:

    gh pr view <PR_NUMBER> --comments
    gh api repos/cloudflare/vinext/pulls/<PR_NUMBER>/reviews
    gh api repos/cloudflare/vinext/pulls/<PR_NUMBER>/comments
    
  2. For each comment:

    • Fix needed: Make the fix in the worktree
    • Out of scope / pre-existing: File a GitHub issue with gh issue create — include context from the review and a link to the PR. Do NOT skip this.
    • Disagree: Reply on the PR explaining why

Read the full file on GitHub · 88 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 88 lines · 18 tokens per session scan A c87b3996c488

Subscribe to this mod's changes

autopilot is a command published in the GitHub repository cloudflare/vinext (8,698 stars, last pushed 3d ago), licensed MIT. It adds 18 tokens to every session and 750 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.