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 saski/arnesto --skill lean-ai-adoption-coachgit clone --depth 1 https://github.com/saski/arnestoWrote 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/saski/arnesto/lean-ai-adoption-coach)<a href="https://agentmods.dev/skills/saski/arnesto/lean-ai-adoption-coach"><img src="https://agentmods.dev/badge/skills/saski/arnesto/lean-ai-adoption-coach/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/saski/arnesto/lean-ai-adoption-coach"><img src="https://agentmods.dev/badge/skills/saski/arnesto/lean-ai-adoption-coach.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 | $0.00076 | $0.00814 |
| Opus 5.5 | $0.00030 | $0.00326 |
| Sonnet 5.5 | $0.00015 | $0.00163 |
| Haiku 4.5 | $0.00008 | $0.00081 |
Grade A, and why
lean-ai-adoption-coach 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.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lean AI Adoption Coach
Goal
Help engineering teams adopt AI in ways that improve flow, quality, learning, and maintainability instead of increasing tool sprawl and orphan assets.
Core Question
Does this reduce waste and improve flow, or just make it cheaper to build more stuff?
Use This Skill When
- The user wants to evaluate a new AI tool, agent, workflow, prompt library, script, or automation.
- The user needs to decide whether a task should be automated or kept manual.
- The user wants guardrails, standards, or rollout guidance for AI usage.
- The user suspects prompt graveyards, agent theater, tool sprawl, or unused AI assets.
- The user wants to know whether an AI initiative creates real value or just more complexity.
Cross-Platform Rules
- Keep recommendations tool-agnostic unless the user explicitly asks about a specific platform.
- Prefer shared patterns over platform-specific one-offs when the same need exists in Codex, Cursor, Claude, Gemini, or Antigravity.
- Use forward-slash paths and portable terminology.
- Penalize any tool, agent, prompt, or automation that has no clear owner or retirement path.
Workflow
- Define the problem in one sentence.
- What friction are we reducing?
- Who experiences it?
- What evidence shows it exists today?
- Classify the work.
- Repetitive and well understood
- Ambiguous but frequent
- High-judgment or high-risk
- Rare or occasional
- Apply the Lean/XP heuristic.
- Favor adoption when the work is frequent, bottlenecked, quickly reviewable, and owned.
- Postpone when the problem is unclear, the tool adds a layer without removing one, or maintenance is ownerless.
- Choose the minimum viable adoption level.
- Guided individual use
- Shared team pattern
- Light workflow integration
- Formal automation
- Cross-team platform or standard
- Define guardrails.
- Owner
- Allowed use case
- Non-allowed use case
- Quality criteria
- Rollback criteria
- Review date
- Retirement path
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.
- 1mo ago First seen · 102 lines · 76 tokens per session scan A 533c1f072faa
lean-ai-adoption-coach is a skill published in the GitHub repository saski/arnesto (6 stars, last pushed 6d ago), licensed Unlicense. It adds 76 tokens to every session and 814 once invoked, about $0.0003 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 2026-09-03.
Other skills, from other repositories
work-state-and-delivery
Use when designing or reconciling design docs, task boards, external trackers, execution plans, delivery records, handoffs, review evidence, task-to-change traceability, or repository commit-coupling policy.
requirements-writer
Use when converting PM intake into a slice SPEC.md with observable acceptance outcomes, explicit scope, and advisory sibling build dependencies.
queue-handoff
Use when ending a turn, finishing a slice, blocked on another agent's work, or escalating to a human — durable work handoff via queue items so the system keeps moving across compactions, missed messages, and interruptions. Covers the hot-potato terminal-turn-rule (active work ends by passing the ball, not by going…
mcore-split-pr
Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.
document-to-action-items
Extract cited facts, obligations, deadlines and proposed tasks from contracts, reports, forms and scans, keeping page citations, how binding each item is (may/should/must) and unclear scan text visible.
issue-triage
Sort a backlog of issues and pull requests: label each by type and severity, link duplicates, try the bug reports, ask for the one missing detail, and turn the accepted ones into ready-to-work briefs.