# Docs

Use this skill when contributing to InsForge's product documentation in this repository.

- Repository: https://github.com/InsForge/InsForge (13274 stars)
- Category: AI and agents
- License: Apache-2.0
- Last updated: 2026-11-01
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add InsForge/InsForge --skill docs -g -a claude-code -y`
Codex: `npx skills add InsForge/InsForge --skill docs -g -a codex -y`
Cursor: `npx skills add InsForge/InsForge --skill docs -g -a cursor -y`
Gemini CLI: `npx skills add InsForge/InsForge --skill docs -g -a gemini-cli -y`
Claude Code plugin marketplace: `/plugin marketplace add InsForge/InsForge`

## Description

Use this skill when contributing to InsForge's product documentation in this repository. This is for maintainers editing public docs in `docs/core-concepts`, agent docs in `.agents/docs`, SDK integration guides in `docs/sdks`, and OpenAPI specs in `openapi`.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# InsForge Dev Docs

Use this skill for `docs/core-concepts/`, `.agents/docs/`, `docs/sdks/`, and `openapi/` in the InsForge repository.

The documentation in this repo is primarily product documentation for InsForge users and agents integrating with InsForge. This skill is for InsForge engineers maintaining that public documentation surface.

## Scope

- `docs/core-concepts/**`
- `.agents/docs/**`
- `docs/sdks/**`
- `openapi/**`

## Working Rules

1. Put each document in the correct documentation surface.
   - Human-friendly docs published on the public doc site belong in `docs/core-concepts/` and related public doc folders.
   - For implementation-heavy public docs, prefer an `architecture.md` file inside the relevant `docs/core-concepts/<domain>/` folder.
   - Agent-only instructions belong in `.agents/docs/`.
   - SDK integration guides for each framework belong in `docs/sdks/`.
   - OpenAPI contract changes belong in the matching files under `openapi/`.

2. Match the writing style to the audience.
   - Public docs should be human-friendly and explain the implementation clearly.
   - Keep public docs human-sounding: avoid AI-writing tells such as em dashes, rule-of-three lists, "not just X but Y" parallelism, inflated significance, vague attribution, and AI vocabulary (delve, leverage, underscore, seamless, robust). See the `doc-author` skill's `INSFORGE.md` overlay, "Sound human, not AI-generated".
   - `architecture.md` pages in `docs/core-concepts/` should explain how the feature works in detail.
   - Agent docs in `.agents/docs/` should be instruction-first and execution-oriented.
   - Agent docs should avoid explanatory filler and focus on the exact steps an agent should follow to complete the work.

The rest: https://github.com/InsForge/InsForge/blob/HEAD/.agents/skills/insforge-dev/docs/SKILL.md