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 axiomantic/spellbook --skill autonomous-modegit clone --depth 1 https://github.com/axiomantic/spellbookWrote 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/axiomantic/spellbook/autonomous-mode)<a href="https://agentmods.dev/skills/axiomantic/spellbook/autonomous-mode"><img src="https://agentmods.dev/badge/skills/axiomantic/spellbook/autonomous-mode/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/axiomantic/spellbook/autonomous-mode"><img src="https://agentmods.dev/badge/skills/axiomantic/spellbook/autonomous-mode.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00106 | $0.02088 |
| Opus 5.5 | $0.00042 | $0.00835 |
| Sonnet 5.5 | $0.00021 | $0.00418 |
| Haiku 4.5 | $0.00011 | $0.00209 |
Grade A, and why
autonomous-mode 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 1mo 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Autonomous Mode
<BEHAVIORAL_MODE> ENTRY GATE: ask, record, then continue with the actual request. Never enable autonomous mode by inference from phrasing alone, and never write the record without the operator's answers. </BEHAVIORAL_MODE>
Invariant Principles
- The record is the mode. Autonomous mode is the file
spellbook/core/autonomous.pywrites, not a statement in the transcript. No record, no enforcement — theStophook allows every turn-end it cannot attribute to a record. So the record is written through the helper below and then READ BACK; an unverified write leaves a mode that is on in the transcript and off in the mechanism. - Ask before recording. Mode and philosophy come from the operator through
AskUserQuestion. Standing autonomous phrasing is the trigger for this question, never its answer. - The goal is captured in the operator's words. The
Stophandler quotes it back on every refusal, so a session that has lost the thread is reminded what it is for. - The exit belongs to the operator. Nothing the agent does ends autonomous mode. Say so when you enable it.
- State the limits when you sell the mechanism. The guarantees below are part of the offer, not fine print disclosed after a failure.
Reasoning Schema
Before asking: state what the operator's request appears to scope, and what would count as its completion. After writing: confirm the record exists, and that the operator has been told the exit and the limits.
The Gate
Ask through AskUserQuestion, in one batch:
Question 1 — how much decision authority?
| Answer | What it means |
|---|---|
fully |
Decisions are yours wherever a reasonable default exists. An unattended decision is APPENDED to the record's decisions list — with the philosophy id active at that moment — instead of announced in the turn. |
mostly |
Genuine forks still come back to the operator through AskUserQuestion. |
What ships with it
1 file 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.
- 1mo ago First seen · 131 lines · 106 tokens per session scan A 9f06dced1498
autonomous-mode is a skill published in the GitHub repository axiomantic/spellbook (11 stars, last pushed yesterday), licensed MIT. It adds 106 tokens to every session and 2,088 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-09-03.
Other skills, from other repositories
batch
Execute batch operations on multiple files in parallel. Automatically discovers files, splits into chunks, and processes with parallel worker agents. Use /batch followed by operation and file pattern. For many independent single-turn transforms (translate/rewrite/extract each file into a new file) that can wait…
batch-api
Prepare a many-file, single-turn transform (translate, rewrite, extract) as a plan and submit it to the asynchronous, half-price DashScope Batch API; results are delivered as new files hours later. Invoke explicitly with /batch-api.
loop
Create a loop that runs a prompt now and follows up either on a fixed schedule or through self-paced wakeups. Usage - /loop check the build, /loop 5m check the build, /loop check the PR every 30m. /loop list to show jobs, /loop clear to cancel all.
openrig-herdr
Use when opening OpenRig fleet terminals as a herdr wall — turning a rig, pod, mission, slice, or saved view into live interactive agent tiles via rig terminal, watching another rig read-only, or driving herdr on an agent's request ("open all my rigs + a mission as views"). Covers the rig terminal open|views|status…
mission-slice-sop
Use when starting, building, handing off, restoring, or closing a mission or slice under the default lightweight SDLC.
community
Use when building or running a persistent two-way community space — Discord, Telegram, Circle: platform choice, structure, onboarding, native→bot→human moderation, rituals, growth loops, health metrics. NOT churn of paying product customers (that is retention), NOT broadcast email (that is newsletter), NOT one live…