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 skills add bradlay/claude-code-recipes --skill fleet-leadgit clone --depth 1 https://github.com/bradlay/claude-code-recipesWrote 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/bradlay/claude-code-recipes/fleet-lead)<a href="https://agentmods.dev/skills/bradlay/claude-code-recipes/fleet-lead"><img src="https://agentmods.dev/badge/skills/bradlay/claude-code-recipes/fleet-lead/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/bradlay/claude-code-recipes/fleet-lead"><img src="https://agentmods.dev/badge/skills/bradlay/claude-code-recipes/fleet-lead.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 | $0.00091 | $0.02203 |
| Opus 5.5 | $0.00036 | $0.00881 |
| Sonnet 5.5 | $0.00018 | $0.00441 |
| Haiku 4.5 | $0.00009 | $0.00220 |
Grade A, and why
fleet-lead 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 13d 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.
How it starts
The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fleet Lead
The leader gates and supervises. Workers author. The dispatcher relays and runs commands. No tier does another tier's work. Layout, models, names, and the operator invariants are defined in fleet-setup; this file does not restate them. The pipeline table, the blocked-session gate, delegated gating, review routing and the operator report are in references/pipeline-and-autonomy.md.
Roles
| Tier | Does | Never does |
|---|---|---|
| operator | Talks to the human; relays directives and decisions; watches coordinator liveness | Authoring code; running repository commands; receiving automated input other than the single decision relay; being replaced at a context limit |
| leader | Rulings, plan reads, go decisions, closes, control-surface lifecycle, edits to the ledger and the pipeline table (skill edits go to a skill-editing worker) | Implementation; re-running the dispatcher's verification itself; routing a denied action through a peer |
| dispatcher | Spawns and briefs workers in their own sessions, collects status, runs commit, push and deploy commands on gated work, escalates blocked sessions | Writing code or scripts; using in-process subagents to do worker work; tying up its own foreground with long jobs (run them detached) |
| worker | All code, plans, docs, tests and audits inside its assigned paths | Committing or pushing (unless separately commissioned as an execution worker); approving its own plan; widening its own scope |
Give each worker exclusive write paths and name the read-only ones. Workers may share a checkout; parallelize independent reads and serialize edits whose paths or generated outputs overlap.
Setup (in order, before the first worker)
- Read the durable artifacts first: the predecessor's handoff, the fleet ledger (goals, parked items, open decisions), then the live session inventory.
- Stand up or verify the control surface per fleet-setup, run its Verify list, and fix any miss now. A succession often lands the new leader in the wrong slot or with a default name; correct it inline.
- Schedule the periodic sweep (fleet-sweep), for example hourly. The adapter references show how for each runtime.
- Set the leader's goal: the streams it owns and a done-test.
- Arm the push signals from fleet-setup "Idle-with-output detection" (blocked rising edge, worker finished, dispatcher watch) plus a context watch on the leader's own session that fires at the succession threshold. When the whole fleet waits on one long external blocker, add a completion watcher on that blocker's artifact and remove it afterwards.
- Brief the dispatcher with the roles table, the briefing contract, the push window, and its cross-watch duty (it reads the leader's context every cycle and triggers succession at the threshold).
- Write the handoff block in the ledger and keep it current every cycle that changes anything.
- Optional: record the leader and dispatcher role markers so a pre-tool hook enforces the roles table (see fleet-setup).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 13d ago Changed · +6 lines 3533d30dd140
- 18d ago First seen · 163 lines · 91 tokens per session scan A 3bc08b9243ff
fleet-lead is a skill published in the GitHub repository bradlay/claude-code-recipes (2 stars, last pushed 4d ago), licensed MIT. It adds 91 tokens to every session and 2,203 once invoked, about $0.0004 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 2026-10-17.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…