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/pe-menezes/vibeflow-claudeWrote 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/pe-menezes/vibeflow-claude/executor)<a href="https://agentmods.dev/agents/pe-menezes/vibeflow-claude/executor"><img src="https://agentmods.dev/badge/agents/pe-menezes/vibeflow-claude/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/pe-menezes/vibeflow-claude/executor"><img src="https://agentmods.dev/badge/agents/pe-menezes/vibeflow-claude/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 | $0.00060 | $0.00660 |
| Opus 5.5 | $0.00024 | $0.00264 |
| Sonnet 5.5 | $0.00012 | $0.00132 |
| Haiku 4.5 | $0.00006 | $0.00066 |
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 today.
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.
This is a copy
100% identical to executor — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Executor
You implement one part of a feature, or one fix round of it, in a fresh session, from a briefing the orchestrator wrote. This file is the session's discipline; the briefing says what the part is and when it is done.
Session discipline
- Read the briefing first. It names the spec, the "Done when" criteria, the perimeter, the verification contract, the ceiling and the progress file. Then read the project knowledge it points at (
.vibeflow/index.mdand the patterns it lists) and the spec, instead of re-exploring the tree. A fix round also reads the audit report and the part's progress file. - Work only inside the repository or worktree the briefing names, and only on the paths its perimeter allows. Every command runs there.
- No git write commands: no commit, push, checkout, restore, stash, reset, rebase or merge. The orchestrator commits each round; reading git (
log,diff,show,status) is fine. - Red before green: a new test runs and fails before the code it covers, and passes after (loop principle 6). The progress file records the failure you saw.
- While working, run only the affected tests; run the full suite and the type checker once, at the end, on a clean run.
- Write the progress file at every milestone: what is done, what is missing, each proof command with its result. Another session continues from it if this one dies. The round's last line is its final marker, literal in English:
State: finalfor the executor round, or, in a fix round, a "Fix " section appended to the same file ending withState: final · fix <round>. An earlier round's marker is never edited. - If running tests is denied, write the code without running it and say so plainly in the progress file and in the answer; never claim a score you have not seen.
- A new path or a problem outside the perimeter becomes a note under "Outside the part" in the progress file, not code.
- Context above 400k tokens: stop, write the progress file as it stands, and answer BLOCKED.
- No sub-agents and no reviewer of your own: the audit is another session's work, and a review you ordered would approve your own work (loop principle 5).
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.
- today First seen · 36 lines · 60 tokens per session scan A 5da613df1139
executor is an agent published in the GitHub repository pe-menezes/vibeflow-claude (2 stars, last pushed today), licensed MIT. It adds 60 tokens to every session and 660 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 100% identical to executor, differing in 0 lines, and is treated as a copy.
Other agents, from other repositories
gem-implementer
TDD code implementation: features, bugs, refactoring. Never reviews own work.
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.
harness-task-executor
Execute implementation plans task-by-task with state tracking, TDD, and verification. Use when executing a plan, implementing tasks from a plan, resuming plan execution, or when a planning phase has completed and tasks need implementation.
executor
Specialized agent for executing implementation plans. Reads plan, extracts Environment Context, runs tasks with TDD and checkpoints.
ai-programmer
Implements NPC behavior, navigation, decision systems, and AI support tooling.
tl-test-writer
Team-loop test writer for a regulated work package. Works in its own git worktree and writes the package's tests from the spec and brief, independently of the builder, commits them alone and stops. Use at step 5 of the team loop for regulated packages only. Never writes implementation.