🌾
AI EngineeringHTML

Harvest

by huytieu

Harvest is an AI Engineering skill for Claude Code, published by huytieu in COG-second-brain.

930 stars110 forkson huytieu/COG-second-brainAdded 2026/08/22+1% in starsRepository updated 2026/08/18
agent-loopagent-pluginsagent-skillsagenticai-agentsclaude-codeclosed-loopcursorgarry-tangbraingstackknowledge-managementloop-engineeringobsidianpeople-crmproductivitysecond-brainself-evolvingverification-harnessworker-agents
Install in seconds
Install Harvest
Copy Harvest into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.
terminal
npx degit https://github.com/huytieu/COG-second-brain/tree/main/skills/harvest ~/.claude/skills/harvest

Requires Node.js. Downloads this skill only β€” not the rest of the repository β€” into your Claude Code skills folder.

Without Node.js

git clone https://github.com/huytieu/COG-second-brain.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file
skills/harvest/SKILL.md in huytieu/COG-second-brain
Installs to
~/.claude/skills/harvest
Collection
One of 25 skills cataloged from this repository
Category
AI Engineering β€” 2451 skills

What Harvest does

Harvest captures session corrections, stable facts, process lessons, and skill gaps for later human review. Use it at the end of a session or from a staging hook when you want to propose durable knowledge without writing it automatically.

Harvest is cataloged under AI Engineering on DirSkills. Harvest comes from a repository tagged agent-loop, agent-plugins, agent-skills, agentic and ai-agents.

Documentation

README

/harvest β€” session learning capture

Adapted from dwarves-kit harvest hook + vault lizard pipeline. Prevents tacit knowledge dying in the transcript.

Triggers

  • /harvest β€” manual end-of-session
  • /harvest promote β€” curate staging β†’ lizard drafts (uses harvest-curator)
  • Staging file written by stop hook (.cursor/hooks/harvest-stager.sh)

Phase 1 β€” Collect (automatic or manual)

Scan the current session for:

  1. Corrections you made ("no, actually X", rejected deliverables)
  2. New stable facts (IDs, URLs, decisions) not yet in 05-knowledge
  3. Process lessons (what worked, what failed, friction)
  4. Skill gaps (something that should be a Verify step or CLAUDE.md rule)

Append to 04-projects/harness/harvest/staging-<YYYY-MM-DD>.md:

This is the opening of the README. Read the full README on GitHub.

Commands Harvest provides

Slash commands named in this skill’s SKILL.md, listed in the order they first appear.

  • /harvest
  • /memory-hygiene

Frequently asked about Harvest

  • What else does huytieu publish alongside Harvest?

    Harvest is one of 25 skills that DirSkills catalogs from huytieu/COG-second-brain, the repository it ships in. Its siblings there include Auto Research, Braindump and Closed-Loop Execute. Each one is a separate skill with its own page in this directory, installs the same way Harvest does, and is maintained by huytieu in that same repository. The rest of the collection is listed on the huytieu/COG-second-brain page.

  • How does Harvest compare to other AI Engineering skills?

    Harvest ranks #1468 by stars among the 2451 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Harvest against them. Open each page to compare what they document and how they install.

More from huytieu/COG-second-brain

Harvest is one of 25 skills cataloged on DirSkills from huytieu/COG-second-brain.

See all 25 skills β†’
πŸ”Ž
1w ago

Auto Research

Auto Research decomposes a strategic question into parallel research threads and synthesizes findings into actionable analysis. Use it for market, competitive, or technology questions that need evidence from multiple sources.
AI Engineering
930110
🧠
1w ago

Braindump

Braindump captures stream-of-consciousness notes and turns them into structured analysis with themes, questions, decisions, and strategic insights. Use it when you want to get ideas out quickly and sort them by domain.
Writing
930110
πŸ”
1w ago

Closed-Loop Execute

Closed-Loop Execute runs a V-model pipeline from plan to build, component verification, integration verification, and acceptance checks. Use it for tasks that need traceable evidence tied to acceptance criteria and post-condition validation.
Quality
930110
πŸ“Š
1w ago

Comprehensive Analysis

Comprehensive Analysis generates a 7-day cross-source review for weekly reviews, board prep, and strategic planning. Use it when you need a full picture of team, product, and delivery health.
Data
930110
πŸ› οΈ
1w ago

Content Factory

Content Factory scouts recent developments in a user’s field, scores them for trend fit and unique angle, and turns selected items into posts, blog drafts, or short videos in the user’s voice. It also enforces deduplication, volume caps, and publish verification.
Automation
930110
πŸ“
1w ago

Create User Story

Create User Story helps draft and create user stories or issues after checking for duplicates across active trackers like Linear, GitHub Issues, or Jira. Use it when you need a tracked feature request with acceptance criteria.
Automation
930110