By LearnPrompt
Extract, compress, and persist agent conversation context into durable memories, then run background proactive checks, split large tasks into phased worker coordination, and verify post-implementation claims for regressions or missing edge cases.
Consolidate recent logs, sessions, and existing memory files into durable topic memories, normalize dates, prune stale entries, and keep MEMORY.md short enough for prompt use.
Run a read-only verification pass after implementation to check whether completion claims are real, validation actually ran, and obvious edge cases or regressions were missed.
Build a lightweight proactive mode with scheduled checks, sleep intervals, concise user briefs, and expiry safeguards so an agent can work in the background without becoming an uncontrolled daemon.
Extract durable memories from recent conversation turns into user, feedback, project, and reference categories while avoiding stale code-state facts.
Compress a long agent conversation into a nine-part continuation summary that preserves request, files, errors, user messages, current work, and the next aligned step.
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
"The parts that separate a fun demo from a stable toolchain: memory, compression, verification, routing, proactive jobs."
Portable agent skills distilled from a publicly mirrored CC coding-agent codebase, then rewritten so they can be installed in Claude Code, Codex, and OpenClaw without depending on private runtime internals.
This repo is not a source dump. It is a cleaned skill pack:
SKILL.md bundlesIf you are building with coding agents, these are the parts that usually separate a fun demo from a stable toolchain:
This repo packages those patterns into six installable skills.
Most agent repos share the same hard problems:
The six skills in this repo package those patterns into download-ready bundles.
Claude Code, Codex, or OpenClaw and wanting a shared skill layerCC-style harness design but needing something publishable and portable| Public name | Slug | What it does | Best hosts |
|---|---|---|---|
| CC Dream Memory | dream-memory | Consolidates logs, recent sessions, and topic memories into a short durable memory index. | Claude Code, Codex, OpenClaw |
| CC Memory Extractor | memory-extractor | Extracts durable user, feedback, project, and reference memories from recent turns. | Claude Code, Codex, OpenClaw |
| CC Verification Gate | verification-gate | Runs a read-only challenge pass after implementation to distinguish verified from merely claimed done. | Claude Code, Codex |
| CC Swarm Coordinator | swarm-coordinator | Splits large work into research, synthesis, implementation, and verification with bounded workers. | Claude Code, Codex, OpenClaw |
| CC Context Compressor | structured-context-compressor | Produces a nine-part continuation summary for long sessions and handoffs. | Claude Code, Codex, OpenClaw |
| CC Kairos Lite | kairos-lite | Adds lightweight proactive jobs with schedule, sleep, brief, and expiry rules. | Claude Code, OpenClaw |
Each bundle lives in skills/ and includes:
SKILL.mdreferences/prompt-template.mdreferences/source-notes.mdscripts/...README.mdIf you only try three things, try these first:
dream-memoryverification-gatestructured-context-compressorThat gives you one memory workflow, one quality gate, and one continuation artifact. It is the fastest way to feel whether this repo is useful.
Run the basic bundle checks first:
bash ./skills/check_all.sh
One-line install (any skills-compatible agent):
# all six skills
npx skills add LearnPrompt/cc-harness-skills -g
# or a single skill
npx skills add LearnPrompt/cc-harness-skills -s verification-gate -g
Or copy one skill into a host runtime manually:
mkdir -p ~/.claude/skills
cp -R ./skills/dream-memory ~/.claude/skills/
Then invoke it in Claude Code:
Use /dream-memory to consolidate the memory directory before we continue.
mkdir -p ~/.codex/skills
cp -R ./skills/dream-memory ~/.codex/skills/
Then invoke it in Codex:
Use /dream-memory and consolidate recent memory files into one stable index.
mkdir -p ~/.openclaw/workspace/skills
cp -R ./skills/dream-memory ~/.openclaw/workspace/skills/
Then check it in OpenClaw:
openclaw skills info dream-memory
Publish to ClawHub after checks:
bash ./skills/publish_all.sh 0.1.0
24小时AI更新雷达。雷达Skill(ai-radar)让Agent零API读取GitHub Pages公开JSON,回答"今天AI圈有什么";伯乐Skill(ai-news-radar)帮你维护信源、抓取脚本、Pages部署。数据来自LearnPrompt/ai-news-radar,GitHub Actions每日刷新。
面向 Agent 的演讲系统:先把资料编排成 AST 大纲和逐页素材决定,再交给下游原生渲染。HTML 路线支持 guizang/frontend-slides/beautiful-html-templates;v1.1 新增 PPT Master 原生可编辑 PPTX、raw .pptx template-fill、speaker notes/转场/native objects 映射和 OOXML 演讲体检。Humanize 不复制模板、不后处理最终 deck;问题写成 fix prompt 回给下游,3 轮封顶。
鲁班 | Luban — 把'能用的Skill'打磨成'能被装、能传播、能验证、能进化'的公共资产。五个动作:验料(挑战前提)·访行(同行对标)·过尺(结构/实测/活体三把尺)·慢刨(冻结基线+验证门)·回炉(发布后迭代)。Battle-tested on a real 1k-star repo.
淘金小镇 | Skillrush Town — 每天从 ClawHub 下载榜 Top100 里淘出新进榜、增速榜和潜力 Skill。每日快照 + 历史对比 + 日报 + 公告板页面,网页是公告板,Skill 是淘金方法。
npx claudepluginhub learnprompt/cc-harness-skills --plugin cc-harnessSkill memory layer for Claude Code — auto-capture, learn, and reuse skills from Acontext
A single-skill package for generating harness blueprints for agentic systems.
This skill should be used when the model's ROLE_TYPE is orchestrator and needs to delegate tasks to specialist sub-agents. Provides scientific delegation framework ensuring world-building context (WHERE, WHAT, WHY) while preserving agent autonomy in implementation decisions (HOW). Use when planning task delegation, structuring sub-agent prompts, or coordinating multi-agent workflows.
Evidence-based agent skills compiler with progressive capability tiers (Quick/Forge/Forge+/Deep).
Harness for Claude Code — skills, /harness:* slash commands, persona subagents, lifecycle hooks, and MCP tools without per-repo `harness setup`. Sibling plugins exist for Cursor, Gemini CLI, and Codex.
Unified capability management center for Skills, Agents, and Commands.