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 jonZlotnik/how-to-agent --skill writing-agent-skillsgit clone --depth 1 https://github.com/jonZlotnik/how-to-agentWrote 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/jonzlotnik/how-to-agent/writing-agent-skills)<a href="https://agentmods.dev/skills/jonzlotnik/how-to-agent/writing-agent-skills"><img src="https://agentmods.dev/badge/skills/jonzlotnik/how-to-agent/writing-agent-skills/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/jonzlotnik/how-to-agent/writing-agent-skills"><img src="https://agentmods.dev/badge/skills/jonzlotnik/how-to-agent/writing-agent-skills.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 | $0.00061 | $0.00890 |
| Opus 5 | $0.00030 | $0.00445 |
| Sonnet 5 | $0.00012 | $0.00178 |
| Haiku 4.5 | $0.00006 | $0.00089 |
Grade A, and why
writing-agent-skills 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.
Writing Agent Skills
A skill has two jobs: fire at the right moment, and change what the agent does once it fires. The description does the first job; the body does the second. Most bad skills fail at one of these — they never trigger, or they trigger and the agent behaves the same as without them.
When to use
- Creating a new skill or reviewing a skill in a PR
- A skill exists but agents never seem to invoke it
- A skill fires but the agent's output looks the same as without it
- Deciding whether guidance belongs in a skill or in an always-on file
Protocol
- Write the description as a routing rule. Lead with a third-person action verb stating what the skill does — "Deletes stale branches", not "Delete stale branches". The description is a catalog entry loaded into every session; an imperative there reads as a standing instruction, while imperatives belong in the body, which loads only on activation. Follow with "Use when" triggers, then a "Do not use" boundary. Keep the first 250 characters self-sufficient — some agents truncate there. State triggers generously: agents under-trigger skills more often than they over-trigger them.
- State each rule with its reason. "Never invent a measurement, because a fabricated number reads as authority" generalizes to cases the skill never mentions; a bare list of MUSTs does not. Use literal language throughout ([[concrete-over-figurative]]).
- Constrain what the agent may invent. Any step of the form "replace vague with specific" or "fill in the details" must name the source of the specifics and say what to do when no source exists: delete the claim, cite where it came from, or ask. A step that demands specificity without a source produces fabrication.
- Give the body three working parts. A worked before/after example in a fenced block; explicit prohibitions with the rationalizations they reject; and a validation loop that tells the agent how to check its output and when to stop.
- Keep the body lean. The description loads in every session, the body loads on activation, and bundled files load on demand — so put reference material in
references/and keep the body under ~500 lines. Cut any rule that does not change behavior. - Test against a baseline. Run an agent on a target task without the skill and record what it does; run again with the skill. If the output does not change, the skill teaches nothing, no matter how well it reads. Validate structure with
skills-ref validateand a quality linter.
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 · 61 tokens per session scan A 40cb996364e8
writing-agent-skills is a skill published in the GitHub repository jonZlotnik/how-to-agent (2 stars, last pushed 1mo ago), licensed MIT. It adds 61 tokens to every session and 890 once invoked, about $0.0003 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
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
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…