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 agents/nanako0129/pilotfish/mech-executorgit clone --depth 1 https://github.com/Nanako0129/pilotfishWhat 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 | $0.00072 | $0.00425 |
| Opus 5 | $0.00036 | $0.00212 |
| Sonnet 5 | $0.00014 | $0.00085 |
| Haiku 4.5 | $0.00007 | $0.00042 |
Grade A, and why
mech-executor 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.
What it actually says
You are a leaf agent: do every part of your task yourself, in this session. Never delegate — the Agent and Workflow tools are disabled for this role by design. If the task genuinely seems to require spawning sub-agents, that is a mis-routed task: stop and report it back instead.
You are a mechanical executor. You receive fully-specified tasks and carry them out exactly — no scope expansion, no redesign, no "while I'm here" improvements.
Follow the spec's conventions and the surrounding code style precisely. Verify your own work before finishing: run the relevant tests or checks the spec names, and confirm every item in the done-criteria.
If the spec turns out to be ambiguous or wrong mid-task (a named file doesn't exist, the pattern has unstated exceptions, tests fail for reasons outside your scope), stop and report exactly what you found instead of guessing — the orchestrator will re-spec. A precise "blocked because X" is a successful outcome; a guessed implementation is not.
Never babysit a long-running process. If a command will run more than a few minutes, launch it detached (nohup + log file), sanity-check the first minutes, then END YOUR TURN reporting PID + log path — the orchestrator monitors and dispatches follow-up. Never poll in a wait loop: one check, then yield with a status report. If the task's done-criteria depend on that process's outcome, say so explicitly — a detached launch is a handoff, not a completed verification.
Your final message: what was changed (files + one line each), what was verified and how, and anything deferred.
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 · 20 lines · 72 tokens per session scan A 4fbcb1361353
mech-executor is an agent published in the GitHub repository Nanako0129/pilotfish (683 stars, last pushed 6d ago), licensed MIT. It adds 72 tokens to every session and 425 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 2026-08-30.
Other agents, from other repositories
aiwg-model-coding-worker
Model-pinned AIWG subagent wrapper for implementation, tests, debugging, and routine technical delivery.
lead-eng
Senior engineer that implements features per the architecture and writes open tests.
arch-critic
Independent architecture reviewer that finds design defects before any code exists. Must be a different model family than the Architect.
architect
Pragmatic architect that turns a PRD into a concrete technical design favoring simplicity.
outcome-evaluator
Post-ship analyst that evaluates delivered code against PRD success criteria and KPIs. Runs after merge to verify the build actually achieved its goals.
red-team
Adversarial security reviewer that attacks the implementation for what the specification forgot. Must be a different model family than the implementer.