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.
npx skills add aliengiraffe/vigilante --skill vigilante-issue-implementation-on-rush-monorepogit clone --depth 1 https://github.com/aliengiraffe/vigilanteWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/aliengiraffe/vigilante/vigilante-issue-implementation-on-rush-monorepo)<a href="https://agentmods.dev/skills/aliengiraffe/vigilante/vigilante-issue-implementation-on-rush-monorepo"><img src="https://agentmods.dev/badge/skills/aliengiraffe/vigilante/vigilante-issue-implementation-on-rush-monorepo/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/aliengiraffe/vigilante/vigilante-issue-implementation-on-rush-monorepo"><img src="https://agentmods.dev/badge/skills/aliengiraffe/vigilante/vigilante-issue-implementation-on-rush-monorepo.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high YARA Match · line 41 YARA rule matched a known malware signature (reverse shell, backdoor, ransomware, C2 framework, or info stealer).Fix: Remove the malware payload or compromised file entirely. Investigate how it entered the skill and audit all other artifacts for additional indicators of compromise.
- high Tool Misuse · line 42 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00062 | $0.01198 |
| Opus 5 | $0.00031 | $0.00599 |
| Sonnet 5 | $0.00012 | $0.00240 |
| Haiku 4.5 | $0.00006 | $0.00120 |
Grade A, and why
vigilante-issue-implementation-on-rush-monorepo 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 12d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Vigilante Rush 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.
Rush Focus
- Read the repo/process context supplied in the prompt before changing code.
- Confirm Rush signals such as
rush.json, project folders, and workspace roots before choosing commands. - Identify the smallest affected package or app scope first and keep that scope explicit in validation and progress updates.
- Prefer Rush-native commands and repo-defined scripts over generic workspace guesses.
- If the affected package or app needs local services, call the bundled
vigilante-local-service-dependenciesskill and prefer the repository'sdocker-compose-launchflow when it exists.
Workflow
- 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.mdwhen presentREADME.md- other root or package-specific docs that affect touched files
- If repository instructions conflict, follow the more specific instruction.
- 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.
- 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 name the likely Rush package or app scope when it is already clear, or state that scope detection is the next step.
- 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 or sub-issue links.
- 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>withgit reset --hard origin/<base>(no commits yet) orgit rebase origin/<base>(commits present). - Use
<base>as the PR target later and mention it in the implementation-plan comment.
- Run
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.
- 12d ago First seen · 71 lines · 62 tokens per session scan A c9a616c0ffda
vigilante-issue-implementation-on-rush-monorepo is a skill published in the GitHub repository aliengiraffe/vigilante (40 stars, last pushed 4d ago), licensed Apache-2.0. It adds 62 tokens to every session and 1,198 once invoked, about $0.0003 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.
Other skills, from other repositories
kano-backlog
Prioritize and refine a GitHub Issues backlog with the Kano model — categorize every open issue as Must-be, Performance, Attractive, Indifferent, or Reverse, apply Kano + priority labels back to GitHub automatically, and recommend the single best next issue to pick up. Use this whenever the user wants to triage…
send-feedback
Explicit Codex workflow: Collect reviewed, sanitized rn-dev-agent feedback and create a GitHub issue only with user confirmation.
pipeline-conductor
Use when a pipeline conductor session is being seeded, or when inspecting/debugging one. Operating procedure for the kirocrew-pipeline-conductor agent - run one issue/PR pipeline on one repository as a supervised fleet. Auto-pick items, preflight every candidate to one deterministic claim verdict, stand up one worker…
babysit
Use when the user says "babysit", "monitor", "keep checking", "keep an eye on", "loop on this PR", "let me know when", or wants polling that outlives a wait+poll window. Same-session monitoring loop for PRs, CI runs, tickets, and deployments using the monitorstart / monitorupdate / autonudgestop MCP tools. The loop…
goal-ledger-conductor
Use when the user hands over a goal too large for one session ("clear the flaky-test backlog", "take this feature from design to PRs", "push these N PRs green") and wants the fleet's state to be readable rather than inferred. Own a long-horizon goal end to end while tracking it in the work ledger - decompose it into…
feature-request
Conversational workflow for gathering user feedback and filing GitHub Issues on the Kiro Crew repository. Load when the user clicks "Request a Feature", wants to report a bug, or suggest an improvement.