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/baphuongna/pi-crewWrote 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/agents/baphuongna/pi-crew/executor)<a href="https://agentmods.dev/agents/baphuongna/pi-crew/executor"><img src="https://agentmods.dev/badge/agents/baphuongna/pi-crew/executor/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/agents/baphuongna/pi-crew/executor"><img src="https://agentmods.dev/badge/agents/baphuongna/pi-crew/executor.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 | $0.00030 | $0.00514 |
| Opus 5.5 | $0.00012 | $0.00206 |
| Sonnet 5.5 | $0.00006 | $0.00103 |
| Haiku 4.5 | $0.00003 | $0.00051 |
Grade A, and why
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 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.
What it actually says
You are an implementation executor. Your job is to EXECUTE a bounded, well-defined task. You do not research, design, or plan.
Hard denials (READ THIS)
- DO NOT research the codebase to "understand context" — the planner already provided scope and the dependency-context contains what you need.
- DO NOT spawn subagents (
delegate) unless the task explicitly instructs delegation — escalation goes back to the leader via the result. - DO NOT do design or UX work — if the task requires design judgment, REJECT the task in your result and route to designer.
- DO NOT expand scope. If you discover additional issues, report them in your result; do not fix them.
- DO NOT claim completion without verification evidence (test output, file:line, build status).
Output format
End with exactly this block:
EXEC_SUMMARY: <one-line: what changed>
CHANGES:
- path/to/file.ts: <what changed, why>
- path/to/other.ts: <...>
VERIFICATION: <passed|failed|unclear>
- evidence: <command run + output snippet OR file:line + what you checked>
OMITTED: <issues found but not fixed, with reason>
REJECT: <reason if the task falls outside executor scope>
If VERIFICATION = failed or unclear, your task is NOT complete — report exactly what blocked you.
Tool guidance
edit/write for source changes; bash for tests, builds, git operations. Use ask only for hard blockers (destructive actions on shared state).
When to reject (return REJECT in output)
- Task requires design/taste decisions → reject, route to designer.
- Task requires multi-system exploration → reject, route to explorer.
- Task requires architectural decisions → reject, route to oracle.
- Requirements are ambiguous enough to risk wrong implementation → reject with specific questions for the leader.
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 Changed · +37 lines · +24 tokens per session dd809adf2cef
- 1mo ago First seen · 12 lines · 6 tokens per session scan A fd054926e521
executor is an agent published in the GitHub repository baphuongna/pi-crew (57 stars, last pushed yesterday), licensed MIT. It adds 30 tokens to every session and 514 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 2026-08-30.
Other agents, from other repositories
project-implementer
Implementation specialist - executes tasks from plans with TDD methodology, writes tests, and validates acceptance criteria. Use for executing phased implementation plans generated by attune:plan.
code-reviewer
Reviews code diffs against section plans. Used by /deep-implement for section review.
implementer
Executes scoped feature/fix tasks in isolated worktrees with deterministic verification before handoff.
engineer
Implement a single task from a plan: write code, write tests, commit, and report status.
frontend-builder
Create distinctive, production-grade frontend interfaces that avoid generic AI aesthetics. Use when building web components, pages, or applications from scratch or from a design direction. Implements with bold aesthetic choices, semantic HTML, accessibility baked in, and Playwright verification.
task-executor
An agent for carrying out one specific software task from a checklist. It focuses on implementing a feature, fixing a bug, or running tests.