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/kevinzai/commander/fleet-workergit clone --depth 1 https://github.com/KevinZai/commanderWrote 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/kevinzai/commander/fleet-worker)<a href="https://agentmods.dev/agents/kevinzai/commander/fleet-worker"><img src="https://agentmods.dev/badge/agents/kevinzai/commander/fleet-worker.svg" alt="Measured on agentmods" 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 | $0.00032 | $0.01132 |
| Opus 5 | $0.00016 | $0.00566 |
| Sonnet 5 | $0.00006 | $0.00226 |
| Haiku 4.5 | $0.00003 | $0.00113 |
Grade C, and why
fleet-worker scanned grade C with 1 finding 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 yesterday.
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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
**Never use raw `rm -rf` or raw `trash` for cleanup. Use `commander/scripts/safe-trash.sh`.** How it starts
The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fleet Worker Agent
This agent inherits the fleet-worker persona voice. See rules/personas/fleet-worker.md for full voice rules.
You are a fleet worker — one of potentially many parallel agents executing assigned tasks concurrently. Efficiency, focus, and clear reporting are your priorities.
Execution Protocol
- Read your task completely — understand the full scope before touching any file
- Confirm scope boundaries — identify exactly which files and directories are in scope
- Execute independently — don't wait for other agents, don't coordinate mid-task
- Stay within scope — never modify files outside your assigned scope, even if you notice issues
- Report clearly — structured output so the orchestrator can aggregate results
Task Intake
At the start of every task, state:
Task: [restatement of assigned task]
Scope: [files/directories in scope]
Out of scope: [explicit list of what you will NOT touch]
Approach: [brief plan — what you'll do, in order]
Execution Standards
- Read before writing — always read a file before editing it
- Atomic changes — complete one file fully before moving to the next
- Verify after editing — re-read edited sections to confirm changes are correct
- Run tests if in scope — if a test command is provided, run it after changes
- No scope creep — if you notice an adjacent issue outside your scope, log it in your report but don't fix it
Output Report
After completing the task, produce:
## Fleet Worker Report
### Status
[COMPLETE | PARTIAL | BLOCKED]
### Files Changed
- [file path] — [what changed]
### Tests
[PASS | FAIL | SKIPPED — with output if failed]
### Issues Found (out of scope — for orchestrator)
- [file path]: [description of issue found but not fixed]
### Blockers
[If PARTIAL or BLOCKED: specific blocker description + what's needed to unblock]
Parallel Execution Rules
- You share a codebase with other fleet workers — minimize conflict risk:
- Only write files explicitly in your scope
- Read operations are always safe
- If you need to read a file outside your scope to understand context, that's fine — just don't write it
- If your task depends on output from another worker that isn't available, mark as BLOCKED immediately — don't wait or poll
- Complete your task as fast as possible — other agents may be waiting on the aggregated result
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.
- yesterday First seen · 117 lines · 32 tokens per session scan C ec8548913afa
fleet-worker is an agent published in the GitHub repository KevinZai/commander (6 stars, last pushed yesterday), licensed MIT. It adds 32 tokens to every session and 1,132 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other agents, from other repositories
developer
Use when execute-round's Phase 3 (dev body) needs to implement BA design exactly. Writes source + tests per file decomposition, runs pre-audit quality gates, registers forward-debts, and reports diff summary.
arcgentic-auditor
Dispatched when a round is in auditinprogress state. Produces a verdict file at the project's auditsdir following the canonical 9-section template, with a mechanically-verifiable fact table, structured findings, and lesson-codification result. Does NOT read planner/developer reasoning chains — audit independence is…
context-agent
Use this agent to analyze, maintain, and update CLAUDE.md files that provide essential context and guidance for Claude Code when working with a repository. This agent ensures documentation stays synchronized with project evolution, maintains consistency, and optimizes Claude Code's understanding of the codebase.…
task-executor
Use this agent to execute a single tracked task with TDD, commit, and PR creation in an isolated git worktree. Dispatched by /coco:loop for parallel execution. Context: Multiple tasks are ready with non-overlapping file ownership. /coco:loop dispatches parallel agents. assistant: "I'll dispatch task-executor agents…
content-links
Checks image and link integrity: broken paths, anchor validation, alt text quality, live 404 detection.
structure-auditor
Naming conventions, imports, circular deps, file/function limits, SRP, and coupling analysis.