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 jrenaldi79/harness-engineering --skill documentationgit clone --depth 1 https://github.com/jrenaldi79/harness-engineeringWrote 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/jrenaldi79/harness-engineering/documentation)<a href="https://agentmods.dev/skills/jrenaldi79/harness-engineering/documentation"><img src="https://agentmods.dev/badge/skills/jrenaldi79/harness-engineering/documentation/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/jrenaldi79/harness-engineering/documentation"><img src="https://agentmods.dev/badge/skills/jrenaldi79/harness-engineering/documentation.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 | $0.00045 | $0.01004 |
| Opus 5.5 | $0.00018 | $0.00402 |
| Sonnet 5.5 | $0.00009 | $0.00201 |
| Haiku 4.5 | $0.00005 | $0.00100 |
Grade A, and why
documentation 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 9d 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 — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project self-documentation
Keep instructions short: instruction file -> documentation index -> relevant document. Use the deterministic scripts for discovery and validation; update explanatory prose from the actual implementation and reviewed evidence.
Locate resources and assess the project
Resolve this loaded SKILL.md's directory as SKILL_DIR. Its scripts/ and
references/ directories are self-contained siblings; do not search another
plugin's cache or assume a fixed user home, shell or install location.
Read the target project's applicable instructions first. Check Node.js is 18+ and inspect the existing documentation configuration, instruction markers and hook arrangement. Choose explicit source roots and a documentation directory. Source inventories read filenames only. Do not include sensitive/private data paths or broaden an existing allowlist without reviewing its scope.
Read metadata and lifecycle rules when classifying or creating documents, migrating a catalog, or resolving validation failures.
Set up documentation only
Before the first installation, create documentation.config.json when a different
generator owns tree/module blocks or a handwritten index must remain intact:
{
"version": 1, "sourceRoots": ["src", "scripts"], "docsRoot": "docs",
"instruction": "CLAUDE.md", "index": "docs/classified-index.md",
"strict": false, "managedMarkers": ["docs"]
}
Use reviewed project roots. This example preserves the handwritten docs/index.md and existing tree/module generator; new projects can use the installer defaults.
Run the bundled installer from the target project with absolute, quoted paths:
node "<SKILL_DIR>/scripts/install.js" --target="<project-root>"
The default instruction file is CLAUDE.md. Select --instruction=AGENTS.md when
appropriate; set --sources=src,lib,scripts and --docs=docs to reviewed roots.
The installer preserves authored text and configuration, copies portable scripts
to scripts/self-documentation/, and generates the configured index. It does not
install package dependencies or edit Claude/Codex settings or permission files.
What ships with it
9 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.
- agents/openai.yaml 227 B
- references/metadata.md 4.8 KB
- scripts/filesystem.js 7.9 KB runs code
- scripts/generate.js 6.6 KB runs code
- scripts/install-hook.js 4.9 KB runs code
- scripts/install.js 8.5 KB runs code
- scripts/metadata.js 4.2 KB runs code
- scripts/migrate.js 2.7 KB runs code
- scripts/render.js 4.9 KB runs code
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.
- 9d ago First seen · 115 lines · 45 tokens per session scan A b3e9ea80aaeb
documentation is a skill published in the GitHub repository jrenaldi79/harness-engineering (69 stars, last pushed 10d ago), licensed MIT. It adds 45 tokens to every session and 1,004 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-07.
Other skills, from other repositories
project-spine-workspace
Use when the user asks about Project Spine hosted workspaces, team sync, shared workspace templates, invites, or commands such as spine login, spine workspace, spine publish, or spine drift check --push.
repository-bootstrap
Audit and establish or upgrade the canonical repository governance baseline. Use only when explicitly asked to bootstrap, standardize, or upgrade a repository; do not use for feature development or source-code restructuring.
document-task
Document completed work — update task log, sync documentation, log observations. Use after finishing a task or batch of changes.
nv-context
Set up state-of-the-art context engineering for any repository. Analyzes codebase, generates multi-level CLAUDE.md/AGENTS.md hierarchy, hooks, session management, and token budgets. Based on 200+ sources including ETH Zurich, Anthropic, Google DeepMind, and Manus production data. For engineers who ship with AI agents.
requirements-writer
Use when converting PM intake into a slice SPEC.md with observable acceptance outcomes, explicit scope, and advisory sibling build dependencies.
post-issue-comment
Post a reply comment on a GitHub issue in English with a natural, concise writing style, based on the intent passed as arguments.