Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Rockielab/rockie-claudenpx agentmods add skills/rockielab/rockie-claude/autopilotWrote 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/rockielab/rockie-claude/autopilot)<a href="https://agentmods.dev/skills/rockielab/rockie-claude/autopilot"><img src="https://agentmods.dev/badge/skills/rockielab/rockie-claude/autopilot/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/rockielab/rockie-claude/autopilot"><img src="https://agentmods.dev/badge/skills/rockielab/rockie-claude/autopilot.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 | $0.00093 | $0.01161 |
| Opus 5 | $0.00046 | $0.00580 |
| Sonnet 5 | $0.00019 | $0.00232 |
| Haiku 4.5 | $0.00009 | $0.00116 |
Grade A, and why
autopilot 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 11d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- autopilot — 92% identical, 8 lines differ
How it starts
The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/autopilot — days-long autonomous operation
This is the "just keep going while I sleep" mode. It composes the pieces you already have (queue, ZCM, journal, calibration, budget, dead-ends, [LEARN]) into a single daemon that can drive an 8×H100 project for a week without human input.
Pre-flight checklist — DO NOT skip
Before starting autopilot, you MUST:
- Populate the queue.
queue.py statusshows ≥ target pending items. - Set a budget.
.claude/budget.tomlexists with project + session ceilings. - Have a launcher.
autopilot.confpoints at aLAUNCHER_CMDthat takes a queue-item JSON on stdin, starts training in the background (nohup + &), and writes its PID + log paths. - Configure ntfy.
NTFY_TOPICis set; you've subscribed in the app. Autopilot will wake you via ntfy on anomalies, ceiling-cross, and cooldown. - A working dry-run gate.
dry_run_gate.sh listshows sentinels for every training script the queue might reference. -
[DEAD-END]registry is current. Any director you've already killed is in the registry — autopilot queue-refill won't re-propose it.
The daemon
nohup bash .claude/scripts/autopilot_loop.sh > .claude/memory/autopilot.log 2>&1 &
Stop:
bash .claude/scripts/autopilot_loop.sh --stop
What the loop does
loop forever:
item = queue.py next --json # atomic claim, $0 LLM cost
if item is None:
if queue_under_target: ntfy(tier=2, "queue empty")
sleep 300
continue
pid = LAUNCHER_CMD < item # launcher backgrounds the job
zcm.sh --pid $pid --log $log # poll, $0 LLM cost, wake on anomaly
match zcm_exit:
0 (clean): leave claimed — agent will close via post-run-review
1 (crash): kill process, increment failures
2 (anomaly): kill process, increment failures
3 (timeout): kill process, increment failures
if consecutive_failures >= max:
ntfy(tier=1, "cooldown, $seconds sleep")
sleep $seconds # anti-burn
seconds = min(seconds*2, max)
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.
- 11d ago First seen · 96 lines · 93 tokens per session scan A 91406db92909
autopilot is a skill published in the GitHub repository Rockielab/rockie-claude (21 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 93 tokens to every session and 1,161 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
jsonld-knowledge-graph
Design and ship a companion JSON-LD knowledge graph (graph.jsonld) next to llms.txt for projects with stable concept-level structure. Encodes domain entities and relationships as schema.org triples for LLM citation. Use when project has matrix / hierarchy / phase-binding structure that prose alone leaves implicit, AND…
llm-as-judge
Design pattern for LLM-as-judge evaluators — binary checks as evidence, one named holistic verdict, no score aggregation. Use when designing or reviewing any LLM-based quality gate, evaluator, judge prompt, or verdict schema; when a judge's rubric scores fluctuate between runs; when you catch yourself asking an LLM…
prompt-perturb
An idea-generation tool that fetches creative prompts from outside sources after removing project-specific context from the search.
constitutional-ai
Anthropic's method for training harmless AI through self-improvement. Two-phase approach - supervised learning with self-critique/revision, then RLAIF (RL from AI Feedback). Use for safety alignment, reducing harmful outputs without human labels. Powers Claude's safety system.
auto-run
Autonomous personalized research loop. Use when the user wants to research a topic autonomously, run a research loop, start adaptive research, or use presets like technique-scout or cross-domain. Triggers on: 'auto run', 'research loop', 'autonomous research', 'run research', 'start research', 'adaptive research'.
ai-evaluation
Systematic evaluation (evals) for LLM and AI products. Design test cases, measure accuracy/quality, track regressions, benchmark models, and build continuous evaluation pipelines. Distinct from traditional software testing with probabilistic outputs.