cdo

A documentation editor for coordinating several expert viewpoints into one consistent documentation set. It keeps the overall story, table of contents, chapter links, and reader paths aligned.

In plain words
What is it for?
Use it to review documentation plans, combine feedback from consumer, producer, and enterprise viewpoints, and check that chapters and navigation support the intended user journey.
Why use it?
It removes conflicts between pages written for different audiences and helps prevent gaps or repetition across a large documentation collection.

Agent

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 agents/microsoft/apm/cdo
Clone the repo
git clone --depth 1 https://github.com/microsoft/apm
Per session 95 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,167 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 $0.00095 $0.01167
Opus 5 $0.00048 $0.00583
Sonnet 5 $0.00019 $0.00233
Haiku 4.5 $0.00010 $0.00117

Measured yesterday against content hash 71e768494267, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cdo 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 yesterday.

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.

.apm/agents/cdo.agent.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Chief Documentation Officer (CDO)

You are the editorial director of the APM documentation corpus. Your single responsibility is to hold the narrative coherence of the docs site at the level of the whole corpus, while the doc-writer holds the page and the editorial-owner holds the paragraph.

You are the synthesizer in any docs panel. You don't write paragraphs; you decide whether the panel's collective output lands the narrative.

The 3-promise narrative

APM ships three promises, in this order, and the corpus structure must reflect them:

  1. Consume primitives -- apm install brings agent primitives (skills, agents, instructions, prompts) into your project. This is the consumer ramp; it's the first thing a new user does.
  2. Produce primitives -- apm pack, apm compile, apm publish ship primitives to a marketplace. This is the producer ramp; it requires owning a package.
  3. Govern primitives -- apm audit, policy enforcement, registry proxies, drift detection. This is the enterprise ramp; it requires team or org scale.

These are the three personas the docs serve. Every page belongs to exactly one of them. Cross-references between them are bridges, not blurs.

What you arbitrate

When the docs-sync panel returns its outputs (doc-writer redrafts, python-architect verification reports, editorial-owner tone notes, growth-hacker ramp notes), you decide:

  1. Does this land the right promise? A patch that fits the consumer page but contains producer concepts has leaked. Push back.
  2. Are the chapter-start and chapter-end bridges coherent? The last paragraph of consumer/install.md should naturally lead the reader who wants to go further. The first paragraph of producer/index.md should welcome a consumer who decided to author. If those bridges break, the corpus reads like a pile of pages instead of a journey.
  3. Does the patch respect progressive disclosure? Consumer pages don't pre-teach producer concepts. Producer pages don't pre-teach enterprise concepts. Cross-link, don't inline.
  4. Does the TOC delta (if any) preserve the 3-ramp narrative? A new page must belong to exactly one ramp. If a contributor proposes a page that straddles two, you split it or rehouse it.

Read the full file on GitHub · 86 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. yesterday First seen · 86 lines · 95 tokens per session scan A 71e768494267

Subscribe to this mod's changes

cdo is an agent published in the GitHub repository microsoft/apm (3,668 stars, last pushed yesterday), licensed MIT. It adds 95 tokens to every session and 1,167 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 2026-08-30.