agentsclimarketplace

Vigilante issue implementation on monorepo

Skill aliengiraffe/vigilante/skills/vigilante-issue-implementation-on-monorepo

Vigilante is a sandbox-first orchestration layer for coding agents. It isolates every task in a git worktree, enforces strict credential scoping, and gives you full audit logs — so your agents can't burn down production.

Install
npx -y skills add aliengiraffe/vigilante --skill vigilante-issue-implementation-on-monorepo

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Implement a GitHub issue end-to-end when Vigilante dispatches work for a watched monorepo. Use the provided worktree, respect repository instructions, comment on the issue as work progresses, and report failures back to GitHub.

SKILL.md

5.6 KB, as published. Nobody here has run it

Vigilante Monorepo Issue Implementation

Overview

Implement one GitHub issue from Vigilante dispatch through validated code changes, a pushed branch, and an opened pull request from the provided worktree. Always work inside the assigned worktree, respect repository instructions, and keep the GitHub issue updated with start, plan, progress, PR, and failure comments.

Monorepo Focus

  • Read the repo/process context supplied in the prompt before changing code.
  • Read the detected monorepo stack and shared local-service contract from the prompt before choosing commands.
  • Limit edits to the packages, apps, or shared modules required for the issue.
  • Prefer targeted validation for the touched workspace scope before broader monorepo validation.
  • Avoid unrelated cross-package refactors unless they are required to complete the issue safely.

Workflow

  1. Inspect issue and repository constraints
  • Read the issue details supplied by Vigilante and confirm the issue scope before coding.
  • Read development constraints from repository markdown files before making changes:
    • AGENTS.md when present
    • README.md
    • other root or area-specific docs that affect touched files
  • If repository instructions conflict, follow the more specific instruction.
  1. Announce session start on GitHub
  • Post a comment on the issue as soon as work begins using vigilante gh issue comment.
  • Include that Vigilante launched the session, the working branch, and that implementation is in progress.
  1. Post an implementation plan early
  • After inspecting the issue and repository constraints, post a concise implementation plan to the issue using vigilante gh issue comment.
  • The plan comment should describe the intended development steps before substantial coding work begins.
  1. Detect a stacked base branch
  • Scan the issue body for an explicit, single-line directive of the form Base branch: <branch-name> (label is case-insensitive; <branch-name> is trimmed and may be surrounded by backticks).
  • Only honor the directive when it appears as its own top-level line in the issue body. Do not infer a stacked base from prose, linked issue numbers, native sub-issue relationships, or branch names mentioned elsewhere in the issue.
  • If no directive is present, branch from and target the watch target's base branch as today.
  • When the directive is present:
    • Run git fetch origin <base> inside the assigned worktree before code changes.
    • If the fetch fails because the branch does not exist on the remote, stop work, post a failure comment on the issue, and do not silently fall back to the default branch.
    • Re-root the work branch onto origin/<base> with git reset --hard origin/<base> (no commits yet) or git rebase origin/<base> (commits present).
    • Use <base> as the PR target later in the workflow and mention the stacked base branch in the implementation-plan comment.
  1. Implement inside the assigned worktree only
  • Use only the provided worktree path.
  • Never edit the root checkout when a worktree was assigned.
  • Keep changes scoped to the issue.
  • Prefer native repository tooling and avoid unnecessary new dependencies.
  • If the affected workspace needs local services, call the bundled vigilante-local-service-dependencies skill and reuse its structured output before creating workspace-specific ad hoc service setup.
  • When containerized local services are still needed after that analysis, use the shared docker-compose-launch contract from the prompt so service startup remains scoped to the assigned worktree.
  1. Validate incrementally
  • Run the most relevant package/app/workspace checks first, then expand only if needed.
  • If validation fails, first inspect the per-issue session log with vigilante logs --repo <owner/name> --issue <n> to determine whether the problem is in the code, test setup, or environment before retrying.
  1. Commit, push, and open a pull request
  • Use vigilante commit for all commit-producing operations. Do not use git commit or GitHub CLI commit flows directly.
  • Commit only issue-relevant changes in the assigned branch.
  • Any commit or amend must preserve the user's existing git author, committer, and signing configuration. Commit on behalf of the user and do not overwrite git config with a coding-agent identity.
  • Do not add Co-authored by: trailers or any other agent attribution for Codex, Claude, Gemini, or similar coding-agent identities.
  • Push the assigned branch to the remote.
  • Open a pull request targeting the repository default branch unless the issue specified a stacked base branch in step 4 or other repository instructions say otherwise.
  • When a stacked base branch was specified, pass --base <base> to vigilante gh pr create and state in the PR body and the PR-opened issue comment that the PR is stacked on <base>.
  1. Report progress and failures clearly
  • Use vigilante gh issue comment for progress updates, milestone updates, PR creation, and execution failures.
  • If execution is blocked, validation fails, or a resumed session is unclear, inspect vigilante logs --repo <owner/name> --issue <n> before retrying or reporting the blocker.
  • A Base branch: directive that points at a branch missing from the remote is a blocker: comment a failure on the issue and stop instead of falling back to the default branch.
  • Keep comments concise, factual, and tied to real progress.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.