session-save
Save or update the current session. Use this skill when asked to handoff, save session, update session, start a new chat, etc. Always call this before using the `update_session` MCP tool directly.
- License 许可
- Apache-2.0
- 最近更新
- 2026-04-27
- Stars
- ★ 6
- Forks
- 2
01 前言 Front Matter
- name
- session-save
- description
- Save or update the current session. Use this skill when asked to handoff, save session, update session, start a new chat, etc. Always call this before using the `update_session` MCP tool directly.
03 SKILL.md 正文
If skipping a step, no need to explain that you did, just skip it.
- Update the session
Call update_session with:
what_happened:
State at end of session, single paragraph. What's true NOW — what got resolved, what's still open, what's blocking the next pickup. One line of texture (grind, breakthrough, spiral) if relevant. Anchor to artifacts — "3 tests in test_banned_words.py fail because X" beats "some tests broke."
If it's longer than 4 sentences, it will be truncated. So be conscise
IMPORTANT DO NOT INCLUDE:
- Anything `git log` knows (diffs, file lists, commit messages).
- Anything that belongs in a memory (preferences, decisions, moments). Don't duplicate.
- Process narration ("we discussed," "we considered," "we decided to").
- Chronology ("first we did X, then Y, then Z").
- Long file paths (use relative file paths)
- Any information that already included in a memory
- Number of files/lines changed
-
for_next_session: Bulleted TODOs, action-verb first, one line each. Each item should be something the next Charlie can pick up and do without re-reading the transcript. "Continue banned-words work" = useless. "Fix 3 failing tests in test_banned_words.py" = actionable. Keep them short and actionable. -
tags: 3–6 noun-form keywords for retrieval. Topics, not actions. Prefer "banned-words-detector" over "improving detection." -
workspace: Current workspace path (absolute).
- Create / Update Memories
Review the session for anything worth remembering. Check activation context and recent recalls first; if a memory already exists for the subject, update it instead of creating a new one. It's fine if there's nothing worth saving — don't invent memories to look productive. If unsure, ask.
04 同仓库更多 Skills
trick
Run Charlie tricks by name or path. Use when the user says "/trick", wants to run a trick, list available tricks, or execute a skill file. Also trigger when the user mentions running a specific trick by name (e.g., "run the session-save trick", "run ship", "do the commit trick").
research
Research a topic and save findings to Charlieverse knowledge. Use whenever the user wants to research something, look up documentation, investigate a library or tool, explore a concept, or asks "what do you know about X". Also trigger when the user says "/research", wants to build up knowledge about a subject, or asks you to "look into" something — even if they don't use the word "research" explicitly.
trick
Run Charlie tricks by name or path. Use when the user says "/trick", wants to run a trick, list available tricks, or execute a skill file. Also trigger when the user mentions running a specific trick by name (e.g., "run the session-save trick", "run ship", "do the commit trick").
research
Research a topic and save findings to Charlieverse knowledge. Use whenever the user wants to research something, look up documentation, investigate a library or tool, explore a concept, or asks "what do you know about X". Also trigger when the user says "/research", wants to build up knowledge about a subject, or asks you to "look into" something — even if they don't use the word "research" explicitly.
copilot
Run a task through GitHub Copilot CLI. Use when the user says "/copilot", wants to delegate work to Copilot, or mentions running something through Copilot. Also trigger when the user wants to use Copilot for a task.
codex
Run a task through OpenAI Codex CLI. Use when the user says "/codex", wants to delegate work to Codex, or mentions running something through Codex. Also trigger when the user wants to use an OpenAI model for a task.