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 PresentJay/autopilot-skills --skill autopilotgit clone --depth 1 https://github.com/PresentJay/autopilot-skillsWrote 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/presentjay/autopilot-skills/autopilot)<a href="https://agentmods.dev/skills/presentjay/autopilot-skills/autopilot"><img src="https://agentmods.dev/badge/skills/presentjay/autopilot-skills/autopilot.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00161 | $0.05975 |
| Opus 5 | $0.00081 | $0.02988 |
| Sonnet 5 | $0.00032 | $0.01195 |
| Haiku 4.5 | $0.00016 | $0.00598 |
Grade C, and why
autopilot scanned grade C with 2 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 6d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- destructive flags: `--force`, `--no-verify`, `git reset --hard`, `rm -rf` outside allow paths Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s --max-time 5 https://api.github.com/repos/PresentJay/autopilot-skills/releases/latest How it starts
The opening of the file, as written. The whole thing — 453 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Autopilot — Self-driving mission runner
You receive a mission from the user and run an autonomous loop until they tell you to stop. The domain is whatever the user defines in the mission file — code self-improvement, doc grooming, research synthesis, ops sweeps, anything that benefits from a tight discover→execute→learn cadence.
This skill replaces ad-hoc loops (ralph-loop, improve-airops) with a domain-agnostic orchestrator: per-project mission, mission-bound governance, idle-aware self-pacing via ScheduleWakeup, immutable milestone trail, self-healing on interruption, and opt-in update notifications.
Boot check
First action on every invocation: locate the mission file.
- Default path:
<repo-root>/.autopilot.log/mission.md. - Fallbacks (legacy / migration):
<repo-root>/.claude/autopilot/mission.md. If found, offer migration to default path on first run. - Outside a git repo:
~/.claude/autopilot/<session-id>/mission.md.
Decision:
- Missing → run "Boot interview" below.
- Exists, user explicitly says "new mission" / "reset" / "redo" → back up to
mission.md.bak.<timestamp>, run interview again. - Exists, normal call or
/autopilot resume//autopilot heal//autopilot status→ enter "Phase 0.5 — Resume + update check", then "Operating cycle".
User signals
| Signal | Effect |
|---|---|
/autopilot |
Auto-detect stale state, then run a cycle (or no-op if mission complete) |
/autopilot resume |
Force Mode: paused → active, run a cycle |
/autopilot heal |
Force interruption recovery (clean working branch, mark crashed cycle, start fresh) |
/autopilot status |
Print current state from state.json + last journal entry (includes installed version line); do not run a cycle |
/autopilot version (or --version, -v) |
Print installed version from frontmatter, latest available from GitHub releases (cached), source URL; do not run a cycle |
/autopilot stop / "멈춰" / "pause" |
Set Mode: paused, no ScheduleWakeup |
/autopilot quick mission="X" (or /autopilot --quick) |
All Q's default-pre-filled, single confirm — new in v1.3.0 |
/autopilot mission="X" risk=L2 cadence=15m ... |
Args parsed into Q1–Q10 pre-fills |
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago First seen · 453 lines · 161 tokens per session scan C ff81faea45ac
autopilot is a skill published in the GitHub repository PresentJay/autopilot-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 161 tokens to every session and 5,975 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
srt-whiteboard-animation
A workflow that turns SRT subtitle files into whiteboard-style hand-drawn animation videos. It plans scenes from the subtitles, draws them in flowing strokes, and renders an MP4.
user-research-cookiy
End-to-end user research assistant — qualitative and quantitative. Use this skill whenever the user mentions user research, user interviews, discussion guides, interview guides, research plans, qualitative research, quantitative research, user surveys, survey design, usability studies, participant recruitment…
seo
Deterministic LLM-first SEO audits for websites, blog posts, and GitHub repositories. Use this when the user asks to "perform SEO analysis", "run SEO audit", "analyze SEO", "check technical SEO", "review schema", "Core Web Vitals", "E-E-A-T", "hreflang", "GEO", "AEO", or GitHub repository SEO optimization. For…
audit-resilience
Audit all external HTTP requests in an open-autonomy agent service for resilience — failure modes, FSM propagation, crash/stuck/side-effect classification, and prioritized fix plan.
bump-versions
Bump open-autonomy and open-aea (and their plugins / upstream-pin tags) to the latest released versions across any Valory downstream repo. Reads HISTORY.md + upgrading.md for both frameworks, applies code changes to dev packages only for OA/OAEA breaking changes, bumps version pins everywhere, then syncs third-party…
playwright-best-practices
Use when writing Playwright tests, fixing flaky tests, debugging failures, implementing Page Object Model, configuring CI/CD, optimizing performance, mocking APIs, handling authentication or OAuth, testing accessibility (axe-core), file uploads/downloads, date/time mocking, WebSockets, geolocation, permissions…