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 TimboGP/timbogp-marketplace --skill lean-rolesgit clone --depth 1 https://github.com/TimboGP/timbogp-marketplaceWrote 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/timbogp/timbogp-marketplace/lean-roles)<a href="https://agentmods.dev/skills/timbogp/timbogp-marketplace/lean-roles"><img src="https://agentmods.dev/badge/skills/timbogp/timbogp-marketplace/lean-roles/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/timbogp/timbogp-marketplace/lean-roles"><img src="https://agentmods.dev/badge/skills/timbogp/timbogp-marketplace/lean-roles.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 | $0.00156 | $0.01160 |
| Opus 5 | $0.00078 | $0.00580 |
| Sonnet 5 | $0.00031 | $0.00232 |
| Haiku 4.5 | $0.00016 | $0.00116 |
Grade A, and why
lean-roles 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 12d 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 — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lean roles — adopt a persona
The plugin's defining move: "one coach, many hats." This skill lets the user put the agent into a specific role so they can co-create, get pressure-tested, or get reviewed — the way they'd use a co-founder, a skeptic, or an advisor. Read ../../reference/roles.md (the role catalog + the shared role-play protocol) before adopting any role.
When to use
The user explicitly wants a persona other than the default Coach — to brainstorm with a co-founder, get attacked by a devil's advocate, or reviewed by a mentor. Also when they ask "what roles can you play?" (explain the cast). If they want a structured customer interview, route to customer-interview; for pitch practice, route to investor-pitch — those run their own scripted protocols.
The roles this skill fronts
| Role | What it does for the user | One-line invocation |
|---|---|---|
| Business partner | Collaborative co-founder — brainstorm variants, share the load, disagree honestly | "let's work on this together" / "be my co-founder" |
| Devil's advocate | Relentless skeptic — pre-mortem the model, attack the riskiest assumption | "poke holes in this" / "try to kill my idea" |
| Mentor | Experienced advisor — review the whole model, prioritize the 1–2 things that matter | "review this as a mentor" / "what would you do next?" |
| Customer | (role-play interviewee) | → use customer-interview |
| Investor | (pitch counterpart) | → use investor-pitch |
The Coach, Strategist, and Analyst are guide roles the agent uses elsewhere by default; this skill is specifically for stepping into a character (Business partner / Devil's advocate / Mentor) on request.
Procedure
- Identify the role. Map the user's phrasing to one of the roles above. If they only said "switch roles" or "what can you play?", list the cast (one line each, from
../../reference/roles.md) and ask which. - Load context. Read
.lean/PROGRESS.md,.lean/canvas.md, and.lean/risks.mdso the persona engages this venture, not a generic one. (For Customer/Investor, hand off to their skills.) - Run the shared role-play protocol (
../../reference/roles.md):- Phase 0 (out of character): confirm the role, what's under scrutiny, and the format; remind them of the
debrief/[square-bracket]break signals; announce entering character. - Phases 1–2 (in character): stay in role per its voice and hard line (Devil's advocate concedes only to evidence; Business partner contributes but won't flatter; Mentor is direct and prioritizes). Don't drop character to coach.
- Phase 3 (debrief, out of character): on the signal, step out and summarize what the scene surfaced — for Devil's advocate, the assumptions that didn't survive; for Business partner, the options generated and the trade-offs; for Mentor, the prioritized next moves.
- Phase 0 (out of character): confirm the role, what's under scrutiny, and the format; remind them of the
- Capture anything load-bearing. If the scene surfaced a new or weakened assumption, mirror it into
.lean/risks.md; log a dated line in.lean/PROGRESS.md. The Mentor's review can be saved alongside the canvas as a dated note. - Hand back / route. Return to the Coach by default, or route to the skill the scene implies (
prioritize-risksafter a devil's-advocate teardown,lean-canvasafter a business-partner brainstorm, a specific experiment after a mentor review).
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.
- 12d ago First seen · 48 lines · 156 tokens per session scan A 5121f4ebc39f
lean-roles is a skill published in the GitHub repository TimboGP/timbogp-marketplace (3 stars, last pushed 2mo ago), licensed MIT. It adds 156 tokens to every session and 1,160 once invoked, about $0.0008 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 2026-08-31.
Other skills, from other repositories
recursive-decomposition
Decompose dense codebase-wide, multi-document, PDF, and aggregation work even when the input fits the context window, following Recursive Language Models (Zhang, Kraska, Khattab, 2025). Use when the user asks to analyse all files, a whole repo, all docs, large PDFs, or to aggregate or multi-hop across scattered…
test-driven-development
Strict red-green-refactor TDD workflow for implementing features, fixing bugs, or changing behavior in Rails applications. Enforces the discipline of writing a failing test before any production code. Use whenever you want to implement with TDD — whether a new feature, a bugfix, a refactor, or any behavior change.
git-commit
Turn the working changes into one or more atomic commits with well-written messages. Use whenever the user runs /git-commit or asks to commit their work, wrap up a feature, or "commit what I have.".
swarm
Run a multi-agent audit of a codebase by spawning specialized parallel subagents (security, performance, tests, architecture, dead-code), then synthesize their findings into a single prioritized action plan. Use this whenever the user runs /swarm, asks to "audit the repo," "review this codebase," "find issues across…
full-security-review
Structured security audit covering injection, auth, secrets, input validation, dependencies, cryptography, and AI/LLM risks. Produces severity-graded findings.
accessibility
Audits code for WCAG 2.2 AA compliance and provides design guidance for accessible components. Covers semantic HTML, ARIA, keyboard nav, contrast, focus management, and motion.