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 robcsaszar/scaffold --skill ai-agents-dot-mdgit clone --depth 1 https://github.com/robcsaszar/scaffoldWrote 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/robcsaszar/scaffold/ai-agents-dot-md)<a href="https://agentmods.dev/skills/robcsaszar/scaffold/ai-agents-dot-md"><img src="https://agentmods.dev/badge/skills/robcsaszar/scaffold/ai-agents-dot-md/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/robcsaszar/scaffold/ai-agents-dot-md"><img src="https://agentmods.dev/badge/skills/robcsaszar/scaffold/ai-agents-dot-md.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 | $0.00108 | $0.02425 |
| Opus 5 | $0.00054 | $0.01213 |
| Sonnet 5 | $0.00022 | $0.00485 |
| Haiku 4.5 | $0.00011 | $0.00243 |
Grade A, and why
ai-agents-dot-md 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 4d 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 — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Create / Review AGENTS.md
Two modes: create a new AGENTS.md or review an existing one against research-backed quality standards.
Mode Detection
- User provides an existing AGENTS.md or points to one → Review mode
- User asks to create/scaffold/generate → Create mode
- Ambiguous → ask:
(c)reate a new AGENTS.md or (r)eview an existing one?
Create Mode
Step 1: Ask placement
Where should the AGENTS.md be placed?
(r)oot — repo-wide baseline
(p)ath — specific directory (provide path)
For monorepos: suggest root + per-package files. Closest AGENTS.md wins.
If AGENTS.md already exists at target path → ask: File exists. (r)eview existing / (o)verwrite / (Q)uit?
Step 2: Expert thinking
Before writing, determine for THIS project:
- What's undiscoverable? What can't be inferred from package.json, configs, or directory structure?
- What's dangerous? What commands or patterns cause silent failures?
- What's ambiguous? Where are there two valid approaches and the agent would guess wrong?
- What's non-standard? What conventions differ from framework defaults?
- What's branch-only? Which instructions apply on every task, and which only on some (releases, migrations, one package)? AGENTS.md is loaded at the start of every session, so each line costs tokens and attention on every task whether or not it applies. Inline what every task needs; push what only some tasks reach into a referenced file behind a pointer. Release steps sitting in a root file are paid for by every bug fix.
If all answers are "nothing" — the project may not need an AGENTS.md. Say so.
For repos without package.json or build configs (docs-only, data repos, simple scripts): focus on Mission + Judgment boundaries only — toolchain and closure sections won't apply.
Step 3: Write in priority order
Add sections ONLY if they pass the undiscoverable test, in this order:
- Mission (2–4 sentences) — project purpose + core constraint the agent can't infer
- Toolchain registry — commands as table, NOT what tools enforce
- Judgment boundaries — NEVER / ASK / ALWAYS tiers
- Closure definition — "Done" = specific exit codes, plus how much the check demands (
every changed model has a migration and a test, notupdate tests) - Escalation rules — what to do when blocked
- Task-organized sections — "When writing… / When reviewing… / When releasing…"
What ships with it
3 files 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.
- 4d ago Changed · +17 lines 8031d9ecef03
- 8d ago First seen · 151 lines · 108 tokens per session scan A d3e3751279f5
ai-agents-dot-md is a skill published in the GitHub repository robcsaszar/scaffold (1 stars, last pushed 5d ago), licensed MIT. It adds 108 tokens to every session and 2,425 once invoked, about $0.0005 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
keep-the-why
Extract and preserve the reasoning code cannot explain - decisions, rejected alternatives, workarounds, incidents, constraints - plus project setup and maintainer interviews. Not for what changed (see Keep a Changelog) - only why.
review
Adversarial fresh-context review of an increment before it ships. Every finding cites path:line and is re-verified. Use when saying "review", "grill this", or "critique the implementation".
sw-do
Implement a SpecWeave increment task by task through the ledger, with evidence per task and a verified close. Use for "implement this", "start working", "continue the increment", "keep going".
done
Close an increment: ledger check, specweave verify, optional review, then specweave complete. Use when all tasks are done and saying "close increment", "we are done", or "finish up".
summarize
A guide for turning conversations, logs, documents, or investigation notes into short, action-oriented summaries. It requires conclusions to be tied to available evidence and distinguishes confirmed facts from items still needing confirmation.
pre-modification-check
Use before modifying, refactoring, moving, or deleting files in a Repowise-indexed repository, especially shared utilities, core modules, public APIs, or files the user did not explicitly identify.