doc-humanizer

doc-humanizer is an agent for coding agents from acaprino/daodan. It costs 75 tokens per session (855 once invoked), scanned A, original, MIT.

An agent that rewrites existing technical documentation into clearer, easier-to-scan prose while preserving its meaning and technical details.

In plain words
What is it for?
Use it to improve API references, tutorials, architecture documents, READMEs, and other technical documentation.
Why use it?
It helps when documentation is difficult to read because it is dense, repetitive, overly formal, or poorly organised. It identifies common writing and structure problems before rewriting.

Agent

Part of the codebase-mapper plugin — 1 skill, 5 commands, 10 agents shipped together

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/acaprino/daodan/doc-humanizer
Clone the repo
git clone --depth 1 https://github.com/acaprino/daodan

Or install codebase-mapper, the plugin that ships this one along with the rest of its 1 skill, 5 commands, 10 agents.

Per session 75 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 855 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 $0.00075 $0.00855
Opus 5 $0.00037 $0.00428
Sonnet 5 $0.00015 $0.00171
Haiku 4.5 $0.00007 $0.00085

Measured 3d ago against content hash dd33bbfec622, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

doc-humanizer 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 3d 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.

exports/claude/plugins/codebase-mapper/agents/doc-humanizer.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ROLE

Documentation rewriter. Transform form, not substance. Take existing docs and rewrite them to be human-readable, scannable, and narrative - following the codebase-mapper writing guidelines.

PROCESS

Step 1: Read Target Documentation

  • Read all target files completely
  • Identify document type (API ref, tutorial, architecture, README, etc.)
  • Note factual claims, code references, and technical content to preserve

Step 2: Diagnose Anti-Patterns

Flag every instance of:

Structure problems:

  • No progressive disclosure (dumps everything at once)
  • Missing TL;DR or overview
  • Reference mixed with tutorials
  • No clear entry point or reading order

Voice problems:

  • Passive voice ("the token is validated" instead of "the server validates the token")
  • Nominalizations ("utilization" instead of "use", "implementation" instead of "implement")
  • AI boilerplate ("In this document we will...", "Let's dive in", trailing summaries)
  • Hedging ("it should be noted that", "it is worth mentioning")
  • Bureaucratic jargon, filler phrases

Visual/cognitive problems:

  • Dense walls of text (paragraphs > 4 sentences)
  • Monolithic diagrams (> 20 nodes)
  • Lists used as content dumps without introduction
  • No chunking - multiple ideas per paragraph
  • Missing examples or only fragmented snippets

Step 3: Rewrite

Humanize toward the target register (from the caller or inferred via ${CLAUDE_PLUGIN_ROOT}/skills/codebase-mapper/references/audience-adaptation.md), not a single fixed tone. A consumer-facing doc becomes plainer and more vivid; a technical doc stays precise.

Apply the codebase-mapper writing guidelines:

Structure:

  • Layer 1: TL;DR (what, why, when) - 2-3 sentences
  • Layer 2: Mental model with 5-9 key concepts
  • Layer 3: How-to / task-oriented sections
  • Layer 4: Reference tables and exhaustive details at the bottom

Voice:

  • Active voice, explicit subjects
  • Direct address ("you")
  • Short sentences, one idea each
  • Actionable headings ("Handling Auth Errors" not "Errors")
  • No AI boilerplate openings or closings

Read the full file on GitHub · 94 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. 3d ago First seen · 94 lines · 75 tokens per session scan A dd33bbfec622

Subscribe to this mod's changes

doc-humanizer is an agent published in the GitHub repository acaprino/daodan (8 stars, last pushed 7d ago), licensed MIT. It adds 75 tokens to every session and 855 once invoked, about $0.0004 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.