write-agent-docs SKILL.md skillA

write-agent-docs from arminzou/dotbrain is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add arminzou/dotbrain/write-agent-docs@git:20261009.2ef2192

Pins git:20261009.2ef2192 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
arminzou/dotbrain
Path
plugin/skills/write-agent-docs/SKILL.md
Versions recorded
4
Latest version
git:20261009.2ef2192, committed Oct 9, 2026
First version recorded
Aug 27, 2026
Size
7,264 bytes
Repository stars
5
License
MIT

Supplies the writing discipline for any document an agent consumes — context pointers, information hierarchy, completion criteria, leading words, pruning. Use when creating or editing AGENTS.md or CLAUDE.md, public project docs agents rely on, Brain context, design or ADR docs, skills, or referenced guidance.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Writing for agents

Use this reference for any document an agent consumes: public project documentation, an
`AGENTS.md` or `CLAUDE.md`, private Brain material, a skill, or a document reached through a
pointer. Their packaging and visibility differ; the writing does not. The goal is predictable
behaviour: the agent takes the same process each run, even when its exact output varies.

When the document is a skill, use the runtime's available skill-authoring capability for packaging,
invocation policy, and validation. Apply the writing principles here to its instructions and
references.

## Context pointers

A **context pointer** is text already in context that names material outside the current context and
states when to read it. A skill description is a pointer. So is an `AGENTS.md` line directing an
agent to an architecture document or private Brain file.

The pointer's wording, rather than its target, determines whether the agent retrieves the material
at the right time. A required document behind a vague pointer is a variance bug: sharpen the
pointer first, and inline the material only when a precise pointer still proves unreliable.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add arminzou/dotbrain/write-agent-docs@sha256:a36f30add5fed26f

The line at the top of this page pins the label git:20261009.2ef2192, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_tqyvdz3ppfq24q5z.svg)](https://markdownregistry.com/a/art_tqyvdz3ppfq24q5z)

1 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 4 versions of this file, between Aug 27, 2026 and Oct 9, 2026, about one every 14 days.

versioncommittedcommitsizeaudit
git:20261009.2ef2192 latest2026-10-09 2ef2192 7,264 BA view · diff
git:20261002.959db852026-10-02 959db85 7,328 BA view · diff
git:20260902.29baf792026-09-02 29baf79 6,606 BA view · diff
git:20260827.670aa602026-08-27 670aa60 6,548 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (7264 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

arminzou/dotbrain · 5 stars · license MIT · pushed 2026-10-22 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_tqyvdz3ppfq24q5z
GET https://markdownregistry.com/api/v1/resolve?ref=arminzou/dotbrain/write-agent-docs
GET https://markdownregistry.com/api/v1/blob/a36f30add5fed26f438c239580d34d4656f870c7464afe1b931b8b7e1e455a7a

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from arminzou/dotbrain

AGENTS.md AGENTS.md file
arminzou/dotbrain · AGENTS.md
git:20261017.5b1062d · audit A · 5 stars
CLAUDE.md CLAUDE.md file
arminzou/dotbrain · CLAUDE.md
git:20260618.b0894a9 · audit B · 5 stars
AGENTS.md@docs AGENTS.md file
arminzou/dotbrain · docs/AGENTS.md
git:20261008.d4dcb14 · audit A · 5 stars
CLAUDE.md@docs CLAUDE.md file
arminzou/dotbrain · docs/CLAUDE.md
git:20261001.90f56cf · audit B · 5 stars
brain-site SKILL.md skill
arminzou/dotbrain · plugin/skills/brain-site/SKILL.md · Set up, maintain, build, and preview a Brain's private site with dotbrain site, including navigation, themes, and…
git:20261009.2ef2192 · audit A · 5 stars
close-design SKILL.md skill
arminzou/dotbrain · plugin/skills/close-design/SKILL.md · Manually closes a shipped, abandoned or superseded design — checks shipment evidence, promotes durable residue, stamps…
git:20261017.5b1062d · audit A · 5 stars
curate-project-context SKILL.md skill
arminzou/dotbrain · plugin/skills/curate-project-context/SKILL.md · Curates a project's public and private agent context. Use when checking context health, reorganizing accumulated…
git:20261017.5b1062d · audit A · 5 stars
dotbrain SKILL.md skill
arminzou/dotbrain · plugin/skills/dotbrain/SKILL.md · The dotbrain operating convention — how a Brain, its work graph, and agent workspaces fit together, and the rules that…
git:20261020.cfed120 · audit A · 5 stars
find-unknowns SKILL.md skill
arminzou/dotbrain · plugin/skills/find-unknowns/SKILL.md · Scans an unfamiliar codebase or feature area against the project Brain and reports the unknown-unknowns and tacit…
git:20261009.1da8791 · audit A · 5 stars
grill-decisions SKILL.md skill
arminzou/dotbrain · plugin/skills/grill-decisions/SKILL.md · Challenges a plan against the project's existing domain model, sharpens its terminology, and writes what settles into…
git:20261017.5b1062d · audit A · 5 stars
iterate-design SKILL.md skill
arminzou/dotbrain · plugin/skills/iterate-design/SKILL.md · Iterates from an active design doc in Goal mode for an explicit automation handoff with a mechanical verifier and hard…
git:20261022.f24169b · audit A · 5 stars
manage-work-graph SKILL.md skill
arminzou/dotbrain · plugin/skills/manage-work-graph/SKILL.md · Maintains the project's private work graph — files, shapes, links, claims, records outcomes for, and closes work items…
git:20261017.5b1062d · audit A · 5 stars

Every file in arminzou/dotbrain

Browse by kind, by grade A, or by owner.