Skip to content

docs: add Agent Skills section for rill init --agent - #9829

Open
nishantmonu51 wants to merge 2 commits into
mainfrom
nishant/agent-skills-docs
Open

docs: add Agent Skills section for rill init --agent#9829
nishantmonu51 wants to merge 2 commits into
mainfrom
nishant/agent-skills-docs

Conversation

@nishantmonu51

Copy link
Copy Markdown
Collaborator
  • Adds a new "Agent Skills" docs section under developers/agent-skills/ with four pages: overview, installation, updating, and a skills reference, documenting the skills generated by rill init --agent
  • Recommends rill init --agent all as the primary command across the docs, with tool-specific options (claude, cursor, agentsmd) as alternatives
  • Cross-links the new section from the Agentic Quickstart, the AI Features guide page, and the external IDE page
  • Fixes a stale rill init --template reference on the IDE page (the flag is --agent)

Checklist:

  • Covered by tests
  • Ran it and it works as intended
  • Reviewed the diff before requesting a review
  • Checked for unhandled edge cases
  • Linked the issues it closes
  • Checked if the docs need to be updated. If so, create a separate Linear DOCS issue
  • Intend to cherry-pick into the release branch
  • I'm proud of this work!


### Existing project

To add agent skills to an existing Rill project without affecting your other files, run `rill init` with only the `--agent` flag:

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[P1] Warn that init overwrites existing agent instructions

When an existing project already has custom AGENTS.md, .claude/CLAUDE.md, or matching generated Cursor rules, the recommended --agent all command replaces them: writeAgentInstructions passes force=true to every initializer, which writes these paths unconditionally. This can destroy existing instructions despite the “without affecting your other files” promise, so overwrite or backup guidance must precede this command; see the true-abstractions rule.


## Entry point

Each format includes an entry point that loads automatically when your agent starts working in the project. It instructs the agent to load the `rill-development` skill before making any changes, so the agent always begins with an understanding of Rill's project structure and workflow.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[P2] Fix the Cursor entry-point guarantee

For --agent cursor, convertToCursorRule creates .cursor/rules/development.mdc, not a skill named rill-development, while the always-applied AGENTS.mdc still tells Cursor to load rill-development. Cursor therefore cannot follow this entry point and is not guaranteed to begin with the development context claimed here; fix the generated reference or document the actual on-demand rule behavior, per the product-accuracy rule.

rill init --agent all
```

The skills work with Claude Code, Cursor, and any agent that supports the `AGENTS.md` convention, such as OpenAI Codex, Gemini CLI, and GitHub Copilot. See the [Installation guide](/developers/agent-skills/install) for setup instructions for each tool.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[P2] Require Agent Skills support alongside AGENTS.md

An agent can support the AGENTS.md instruction-file convention without implementing Agent Skills discovery under .agents/skills; in that scenario it reads the generated entry point but cannot resolve the named rill-development skill, so the detailed instructions never activate. Limit this claim to agents verified to support both conventions, or make the generated AGENTS.md self-contained, as required by the true-abstractions rule.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants