MoonshotAI / MoonshotAI/kimi-code
Add a bundled Kimi Code docs system skill
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 7.5k
- Forks
- 1.2k
- Avg merge
- 11h 53m
- Merged PRs (30d)
- 350
Description
Problem:
Kimi Code has built-in skill commands, but there is no bundled product-docs skill comparable to Codex's system product-docs help. When users ask about Kimi Code setup, config, MCP, skills, plugins, slash commands, sessions, IDE/ACP, or troubleshooting, agents do not have an explicit Kimi Code self-help route.
Proposal:
Add a System skill scope for product-aware bundled skills. Include a kimi-code-docs system skill with guidance to consult official docs, local docs, and source, and expose it as /kimi-code-docs. Proposed priority: Project > User > Extra > System > Built-in.
Implementation direction:
Use Kimi's existing in-memory bundled skill registration style rather than unpacking files to KIMI_CODE_HOME. The system skill is registered as system://kimi-code-docs at session startup after project/user/extra roots load and before built-in skills register.
Validation from prototype:
- agent-core/protocol/kimi-code typecheck pass
- targeted skill/TUI vitest pass
- docs build pass
- protocol and agent-core build pass
I have a draft implementation ready and can open a PR if this direction is acceptable.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by tracing Kimi's existing in-memory bundled skill registration and session-startup ordering, including the agent-core/protocol/kimi-code area mentioned in the validation notes. Confirm the system://kimi-code-docs skill and /kimi-code-docs command are exposed with the proposed priority, then run the targeted skill/TUI vitest checks, typecheck, docs build, and protocol and agent-core builds.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- cli, tooling
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100