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.
git clone --depth 1 https://github.com/Ertinox7711/SGRR-AGI-V2Wrote 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/commands/ertinox7711/sgrr-agi-v2/autonomous)<a href="https://agentmods.dev/commands/ertinox7711/sgrr-agi-v2/autonomous"><img src="https://agentmods.dev/badge/commands/ertinox7711/sgrr-agi-v2/autonomous/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/commands/ertinox7711/sgrr-agi-v2/autonomous"><img src="https://agentmods.dev/badge/commands/ertinox7711/sgrr-agi-v2/autonomous.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00028 | $0.00551 |
| Opus 5.5 | $0.00011 | $0.00220 |
| Sonnet 5.5 | $0.00006 | $0.00110 |
| Haiku 4.5 | $0.00003 | $0.00055 |
Grade A, and why
autonomous 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 21d 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.
What it actually says
Bounded autonomous mode for the current project. Port of prime-agent's
/autonomous, driven by the Stop hook.
Arguments: $ARGUMENTS
Engine: python ~/.claude/scripts/autonomous.py --cwd <project root> <sub>
| input | run |
|---|---|
on <objective> |
on "<objective>" plus any --gate / budget flags given |
off |
off - stop continuing immediately |
(nothing) or status |
status - active?, continuations spent, gates, budget |
gate-add <cmd> / gate-rm <cmd> / gate-clear |
the matching subcommand |
budget --max-turns N ... |
budget with the flags given |
Defaults, same as upstream: 3 continuations, 12 turns, 80 000 tokens, 30 min,
per-gate timeout 120 s, 2 gate retries. Any limit accepts unlimited; pair an
unlimited budget with a gate or it only stops on error. Tokens counted are new
tokens (input + output + cache creation) since activation - cache reads excluded.
How it behaves once on:
- With gates: at the end of each turn every gate runs. All green -> autonomous switches itself off and the turn ends. A gate fails -> the failure output is injected and work continues. A gate that keeps failing with an unchanged workspace stops the run instead of burning budget on the same failure.
- Without gates: work continues until a budget is spent. Reaching a limit never means the task succeeded - say what is actually done.
Rules:
- Turn it on only when the operator asks. Never infer it.
- Never bypass a failing gate (
--no-verify, skipping a test, disabling a check) and never remove anything to make one pass - that is the whole point of the gate. If the gate itself is wrong, stop and say so. - When the work is genuinely finished, or blocked on an input only the operator can
give, run the
offsubcommand and then stop.
After any state change, report the new status in one line.
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.
- 21d ago First seen · 46 lines · 28 tokens per session scan A ac40ac0d3084
autonomous is a command published in the GitHub repository Ertinox7711/SGRR-AGI-V2 (1 stars, last pushed 1mo ago), licensed MIT. It adds 28 tokens to every session and 551 once invoked, about $0.0001 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 2026-10-04.
Other commands, from other repositories
prototype
You are building a proof-of-concept for the current Grainulator sprint. Read CLAUDE.md for sprint context and claims.json for existing research claims.
qa-changes
This skill should be used when the user asks to "QA a pull request", "test PR changes", "verify a PR works", "functionally test changes", or when an automated workflow triggers QA validation of code changes. Provides a structured methodology for setting up the environment, exercising changed behavior, and reporting…
verify
Run repository verification using the verification-loop skill.
modernize-transform
Transform one legacy module to the target stack — idiomatic rewrite with behavior-equivalence tests.
constraints
A command for defining a project's quality standards through a short interview and a written CONSTRAINTS.md file. It records checks, thresholds, exceptions, and commands for enforcing them.
check-dev
Type-check a Z specification with fuzz.