MoonshotAI / MoonshotAI/kimi-cli

Feature Request: Add default_skills config for auto-activating skills on session start

Open
#2,062 2 comments 2 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
11.4k
Forks
1.3k
Avg merge
9h 47m
Merged PRs (30d)
2

Description

What feature would you like to see?

A config option to automatically activate specific skills when a new session starts, without requiring the user to manually trigger them via slash commands.

Motivation

Users who rely on specific skills for every session (e.g., communication style skills like caveman mode, or workflow skills like git-commits) currently have to remember to type the slash command every time they start Kimi. This is repetitive and easy to forget.

Proposed Solution

Add a top-level default_skills key to config.toml:

default_skills = ["caveman", "git-commits"]

Behavior:

  • On session start, Kimi reads this list and auto-invokes each skill as if the user had typed /skill:<name>.
  • If a skill name does not exist, log a warning and skip it (do not fail startup).
  • Default value is an empty list [], so the change is fully backward compatible.
  • Works alongside existing AGENTS.md and system prompt injection — skills are applied after system prompt setup but before the first user turn.
Scope

This request is intentionally minimal: just automatic activation of already-discovered skills via config. It does not cover skill auto-discovery paths (already solved by extra_skill_dirs in #2044) or rules systems.

Additional information

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by locating the config.toml loader, session-start entry point, and existing /skill: handling. Trace how skills are discovered and invoked alongside AGENTS.md and system prompt setup; done means applying configured skills before the first user turn, warning and continuing for unknown names, and preserving the empty-list default.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
cli
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Mostly clear
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.