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/inferencegod/autonomy-loop/autonomy-operatenpx skills add inferencegod/autonomy-loop --skill autonomy-operategit clone --depth 1 https://github.com/inferencegod/autonomy-loopWhat 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 | $0.00062 | $0.01409 |
| Opus 5 | $0.00031 | $0.00705 |
| Sonnet 5 | $0.00012 | $0.00282 |
| Haiku 4.5 | $0.00006 | $0.00141 |
Grade A, and why
autonomy-operate 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.
How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
autonomy-operate - the rhythm
Read autonomy.config.json for this repo's knobs. Read STATE.md first every session.
The non-negotiables
- The baton is truth. Work only when
LOOP-STATE.mdturn:is your role. Never infer the turn from the commit log. - Write everything down. STATE.md / LOOP-STATE.md / REVIEW-FEEDBACK.md hold the thread - chat compacts and is lost. Never rely on memory.
- The gate, every wave:
gate.test+ the frozen invariant intact (gate.frozenInvariant) +gate.build+gate.linton touched files. Ifgate.coverageis set, also runnode ${CLAUDE_PLUGIN_ROOT}/hooks/coverage-ratchet.mjsafter the coverage command: a coverage drop below the.autonomy-coverage.jsonfloor is a RED gate (the third gate, the drift guard). Ifgate.patchTargetis greater than 0, also runnode ${CLAUDE_PLUGIN_ROOT}/hooks/patch-coverage.mjs --threshold=<patchTarget>so the wave's own changed lines are tested (the fourth gate; the reviewer passes--base=<last-reviewed-sha>). All green or REVERT - never commit red. Ifgate.envFidelityis set, the LOCAL gate must match CI (an env-naked pass is not proof CI is green). - Honesty mandate (
honestyRule): no fabricated numbers; every rate carries N + CI or "building - N/30"; a capability with no real data ABSTAINS visibly - never faked, never silently dead. - Gate List (
gateList): both terminals STOP and ask the human for risky/irreversible actions. Never self-authorize one - escalate toFOR-REVIEW.md, setturn: human. - Push the work branch only, never prod. (The gate-guard hook enforces this; don't fight it.)
Builder rhythm
Two modes - IDEATE (research lane) and BUILD. ultrathink the plan. Smallest high-signal diff; PURE module + RED-before-green test for new logic; zero deletions on shared files; re-read fresh. Pre-empt the reviewer by answering the 5-lens rubric (Correctness · Honesty · Regression/frozen-drift · Security · UX) in your build note, with a "red-team the opposite" paragraph.
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.
- yesterday First seen · 44 lines · 62 tokens per session scan A ac60376e2823
autonomy-operate is a skill published in the GitHub repository inferencegod/autonomy-loop (1 stars, last pushed 2mo ago), licensed MIT. It adds 62 tokens to every session and 1,409 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 2026-08-31.
Other skills, from other repositories
hive.chart-creation-foundations
Required reading whenever any chart tool is available. Teaches the one-tool embedding contract (call chartrender → live chart appears in chat AND a downloadable PNG lands in the queen session dir), the ECharts (data viz) vs Mermaid (structural diagrams) decision, the BI/financial-grade aesthetic baseline (no…
browser-edge-cases
SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.
openai-whisper-api
Transcribe audio via OpenAI Audio Transcriptions API (Whisper).
taiyi-ui-design
TaiyiForge 第 4 阶段 — UI/UX 契约，产出 UI-DESIGN.md。四端通用。.
taiyi-evolve
TaiyiForge 辅助 — 实现后架构与文档同步（architecture-sync）。OpenCode / Claude / Codex / Cursor 通用。.
perf-code-paths
Use when mapping code paths, entrypoints, and likely hot files before profiling.