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 agentmods add skills/pekral/cursor-rules/autonomous-loopsnpx skills add pekral/cursor-rules --skill autonomous-loopsgit clone --depth 1 https://github.com/pekral/cursor-rulesWrote 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/pekral/cursor-rules/autonomous-loops)<a href="https://agentmods.dev/skills/pekral/cursor-rules/autonomous-loops"><img src="https://agentmods.dev/badge/skills/pekral/cursor-rules/autonomous-loops.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 | $0.00089 | $0.02075 |
| Opus 5 | $0.00044 | $0.01038 |
| Sonnet 5 | $0.00018 | $0.00415 |
| Haiku 4.5 | $0.00009 | $0.00208 |
Grade A, and why
autonomous-loops 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 3d 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 — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
This is a reference skill. It does not run a loop itself — it helps you pick the right autonomous-loop architecture for a task and wire it to this project's existing skills, gates, and CLI tooling. Patterns are ordered simplest → most complex. Use the lightest one that fits.
Constraints
- Apply
@rules/git/general.mdc— never push tomain, one logical change per commit, English commit messages, worktree/branch per work unit. - Apply
@rules/code-review/general.mdcto every review stage in a loop — reviewers are read-only and must not be the same context that wrote the code. composer buildandcomposer skill-checkare non-negotiable gates between iterations. An iteration that does not pass them is not "done" and must not advance the loop or merge.- Every loop needs an explicit stop condition (max iterations, no open work, or a documented blocker). Never run unbounded.
- Stop on blockers, never force through them: merge conflict, failing CI, unresolved Critical/Moderate findings, or a gate failure ends the loop with a report.
- Parallel multi-unit patterns (e.g. the DAG below) are an explicit opt-in: only when the user chooses parallel orchestration does each work unit run in its own git worktree so units cannot corrupt each other's tree. A single sequential loop never creates a worktree on its own — it works in the current tree per
@rules/git/general.mdcWorktrees / Workspaces. - Do not expose sensitive/internal details in user-facing loop reports.
Use when
- Deciding whether a task should run autonomously at all, and with which architecture.
- Setting up a scripted or scheduled pipeline that chains this project's skills without a human between steps.
- Running multiple independent work units in parallel and coordinating their merges.
- You need context to survive across otherwise-independent iterations.
When NOT to run autonomously
- The requirements are vague or the acceptance criteria are missing — autonomy multiplies hallucination. Resolve scope with a human first (see
prepare-issue-context). - The change touches auth, payments, migrations on production data, or anything irreversible.
- One focused edit a human could do faster than wiring a loop.
- No reliable automated gate exists for the work (nothing for
composer build/ tests to verify).
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.
- 3d ago First seen · 132 lines · 89 tokens per session scan A 88c5a06d0842
autonomous-loops is a skill published in the GitHub repository pekral/cursor-rules (6 stars, last pushed 3d ago), licensed MIT. It adds 89 tokens to every session and 2,075 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
browse-and-evaluate
Use when exploring the ai-agent-skills catalog to find, compare, and evaluate skills before installing. Always use --fields to limit output size and --dry-run before committing to an install.
loop-engineering
Shared loop-engineering reference for COG skills - the agent loop, deterministic verifiers, termination conditions, in-loop context management, and named patterns. Invoke when designing or debugging a skill that iterates (search-verify-retry, scan-until-dry, fetch-retry-gate).
render-airdrop-carousel
Assemble a viral iOS "AirDrop" notification-carousel video ad (≈6–8s, 9:16) from a brand line plus 6–16 real product photos — a native AirDrop share-sheet card ("Brand would like to share a · Decline / Accept") springs up and its preview window CYCLES through the products, landing on a range/lineup payoff with an…
render-3d-product-showcase
Assemble a premium 3D product-showcase ad from a config — four beat clips (an orbiting hero rotation, a macro push-in, a physics reveal, a typographic close) normalized to the brand-color canvas, hard-concatenated in order, closed on a deterministic Playwright brand end card, and mixed under one instrumental bed at…
vfx-text-cursor
Cursor light trail, chromatic rays, and directional flares for word-by-word quote reveals in video intros.
music
Generate, remix, extend, edit, and analyze AI music (Mureka). Triggers on: "音乐", "music", "生成音乐", "generate music", "翻唱", "cover", "混音", "remix", "续写", "extend", "纯音乐", "instrumental", "配乐", "soundtrack", "分轨", "stem", "识别歌词", "recognize lyrics", "作曲", "compose", "create a song", "做一首歌".