writing-agents-config

A guide for creating and reviewing agent rules and skills in a repository. It explains how always-on rules, file-specific rules, and task-triggered skills should work together.

In plain words
What is it for?
Use it when adding, organizing, or checking `.agents/` rules and skills, including deciding which instructions belong in each kind of configuration.
Why use it?
It prevents duplicated instructions, oversized rules, and unclear links between rules and skills. It also helps keep the repository’s agent configuration consistent.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/stainless-code/codemap/writing-agents-config
Any agent
npx skills add stainless-code/codemap --skill writing-agents-config
Clone the repo
git clone --depth 1 https://github.com/stainless-code/codemap

Made for: Claude Code, Codex.

Per session 54 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,416 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00054 $0.01416
Opus 5 $0.00027 $0.00708
Sonnet 5 $0.00011 $0.00283
Haiku 4.5 $0.00005 $0.00142

Measured 2d ago against content hash cbec72e15670, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

writing-agents-config 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 2d 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.

.agents/skills/writing-agents-config/SKILL.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Writing agents config

Read first: writing-great-skills — invocation, information hierarchy, completion criteria, pruning, failure modes.

This repo adds Cursor rules (Tier 1/2/3) on top of a skills-only agent model — always-on STOP rules where cross-turn non-negotiables beat description-only triggers.

Tier framework: agents-tier-system · Layout: agents-first-convention.

Attachment decision tree

  1. Every turn, non-negotiable? → Tier 1 (alwaysApply: true) — ≤40 lines STOP + pointers
  2. File-scoped pattern? → Tier 2 (globs:) — thin priming; depth in paired skill
  3. Intent-only workflow? → Tier 3 or skill-only — no always-on tax (bug diagnosis, doc sweeps, structural queries)
  4. User slash-command only?disable-model-invocation: true wrapper → model skill(s)

Pairing checklist

Check Pass when
Priming size Rule ~10–40L; procedures in skill or sibling
Bidirectional Rule → skill in Reference; skill → rule where paired
No duplication SSOT in skill; rule has STOP rows or pointers only
Inventory drift Tier-1 set discovered via the frontmatter audit in agents-tier-system rule — not hardcoded name lists in README/skill
Triggers Per writing-great-skills — one branch per intent in description

SKILL.md size tiers

Rules cap at ≤40L; skills use progressive disclosure (writing-great-skills sprawl). Line count is a hygiene signal, not a hard gate — split for structure, not to hit a number.

Read the full file on GitHub · 81 lines

Changes

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.

  1. 2d ago First seen · 81 lines · 54 tokens per session scan A cbec72e15670

Subscribe to this mod's changes

writing-agents-config is a skill published in the GitHub repository stainless-code/codemap (8 stars, last pushed 7d ago), licensed MIT. It adds 54 tokens to every session and 1,416 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.

Related

Other skills, from other repositories