跳到正文
cccskills
免费在 GitHub 公开

m0

Use when the user says 'm0', 'm0 status', 'start m0', 'cross-tool memory', 'operational thread', or 'where is my memory stored', or when the M0 daemon or MCP wiring needs attention.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(3)

  • SKILL.md3.0 KB
  • scripts/m0_mcp.py15.3 KB
  • scripts/m0_server.py23.6 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

/m0 — Cross-Tool Agent Memory

M0 keeps one local operational thread per project: what was done, what was verified, what is next. Any tool can write to it and read it back, so a session that starts cold can continue work another tool began.

The live M0 is the Rust daemon coco-m0d, serving 127.0.0.1:8000 under the launchd job com.coco.m0-backend. The write path is /m0-remember, the read path is /m0-recall, and the session handoff flow is /m0-handoff.

The token is read from ~/.coco/m0-token on each call; never print it.

Quick Reference

M0="${COCO_M0_URL:-http://127.0.0.1:8000}"     # daemon base URL

# Is it up? (health needs no token)
curl -s "$M0/api/health"

# Is the launchd job loaded?
launchctl list | grep com.coco.m0-backend

# Everything else needs the token, for example the recent thread:
curl -s -H "Authorization: Bearer $(cat ~/.coco/m0-token)" \
  "$M0/api/brain/thread?project=my-project&limit=5"

There is no offline path: if the daemon is down, say so and stop; do not write anywhere else.

Sub-commands

/m0 status — is memory working

M0="${COCO_M0_URL:-http://127.0.0.1:8000}"
curl -s "$M0/api/health"
launchctl list | grep com.coco.m0-backend

Report to the user, in this order:

  1. Daemon — reachable at $M0 or not.
  2. launchd — whether the job com.coco.m0-backend is loaded and running.

If either is missing, tell the user and stop. Do not work around it.

/m0 start — you do not start it

The daemon runs under launchd; if it is down, tell the user and stop (restarts and deploys are the owner's).

/m0 mcp — the MCP tool

The daemon's MCP binary coco-platform-mcp exposes one tool, m0, with modes recall, context, wip, handoff, and remember. Call it directly — it is the lowest-friction way to make memory habitual.

/m0 verify — prove it works

curl -s "$M0/api/health"

A healthy response means the daemon is serving. Quote the output rather than summarising it.

/m0 privacy — what leaves the machine

Nothing. The daemon binds to loopback, there are no outbound calls in the request path and no telemetry. Text fields are capped at 400 characters and redacted by the daemon before storage.

When to reach for this

  • Use /m0-recall at the start of a session, or when picking up work started elsewhere, before asking the user to repeat context.
  • Use /m0-remember after a step lands, a decision is made, or something is verified.
  • Use /m0-handoff before context is compacted or a session ends.
  • Use /m0 (this skill) for the plumbing: daemon, token, wiring.

M0 answers "where were we and what is next". For deeper knowledge-graph retrieval see the comparison with the cognee bundle in systems/m0/README.md before choosing.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Train and optimize AI agents using Microsoft's Agent Lightning framework with reinforcement learning. Use when setting up agent training, instrumenting agents with tracing, configuring LightningStore, implementing reward functions, or optimizing prompts with RL/APO algorithms.

中文简介正在准备中。当前显示原文说明。

coco-research/coco7922026年10月19日 更新

Post-run self-evaluation system that scores agent output on correctness, clarity, actionability, and conciseness. Use after /team runs, skill executions, or when explicitly asked to evaluate output quality.

中文简介正在准备中。当前显示原文说明。

coco-research/coco7922026年10月19日 更新

Create AI marketing videos for ads, promos, product launches, and brand content. Models: Veo, Seedance, Wan, FLUX for visuals, Kokoro for voiceover. Types: product demos, testimonials, explainers, social ads, brand videos. Use for: Facebook ads, YouTube ads, product launches, brand awareness. Triggers: marketing video, ad video, promo video, commercial, brand video, product video, explainer video, ad creative, video ad, facebook ad video, youtube ad, instagram ad, tiktok ad, promotional video, launch video

中文简介正在准备中。当前显示原文说明。

coco-research/coco7922026年10月19日 更新

Use when building AI features into a product: LLM integration, RAG pipelines, guardrails, streaming, AI UX, prompt engineering, or AI cost control. Treats prompts as code and validates every model output.

中文简介正在准备中。当前显示原文说明。

coco-research/coco7922026年10月19日 更新

Your AI research and engineering brain trust. 59 named personas across 8 cells covering frontier labs, applied product, model architecture, reasoning/RL/agents, alignment and interpretability, theory and science of DL, multimodal and…

中文简介正在准备中。当前显示原文说明。

coco-research/coco7922026年10月19日 更新

Use when designing a new REST or GraphQL API, reviewing an API spec before implementation, setting team API standards, or migrating REST to GraphQL. Covers resources, HTTP semantics, pagination, error handling, and pitfalls.

中文简介正在准备中。当前显示原文说明。

coco-research/coco7922026年10月19日 更新

查看 coco-research 的全部技能

报告此技能的问题