abhattacherjee

Claudeception — Development skill for Claude Code

Development community

Extracts reusable knowledge from work sessions and codifies it into Claude Code skills.

How to install Claudeception

This entry records only its repository, not the path inside it, so there is no exact command to give. Open abhattacherjee/claudeception and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Claudeception does

Extracts reusable knowledge from work sessions and codifies it into Claude Code skills.

Alternatives in Development

  • Knowledge Work Plugins — Open source repository of plugins primarily intended for knowledge workers to use in Claude Cowork 9734 125 18 10.5k ★
  • Resume Work From A Handoff Document — You are tasked with resuming work from a handoff document through an interactive process 10k ★
  • Agent Rules — Rules and Knowledge to work better with agents such as Claude Code or Cursor 5.7k ★

README

Claudeception

A customized fork of [blader/Claudeception](https://github.com/blader/Claudeception) — a continuous learning system that extracts reusable knowledge from work sessions and codifies it into Claude Code skills.

Every time you use an AI coding agent, it starts from zero. You spend an hour debugging some obscure error, the agent figures it out, session ends. Next time you hit the same issue? Another hour. This skill fixes that by persisting discoveries as reusable skills.

Changes from Upstream

This fork customizes the original Claudeception with:

  1. Skill-authoring integration — delegates skill creation to the skill-authoring skill for consistent structure, frontmatter conventions, and quality checks
  2. Existing skills check (Step 1) — before creating a new skill, searches project and user-level skill directories to decide: create new, update existing, or add cross-references
  3. Project artifact updates (Step 5) — after saving a skill, updates CHANGELOG.md and uses appropriate commit prefixes (docs(skills): or fix(skills):)
  4. Streamlined frontmatter — uses metadata.version instead of top-level version, removes author, date, tags, and allowed-tools fields per Anthropic best practices
  5. Concise research section — replaced verbose search instructions with a one-liner strategy
  6. See Also cross-references — links to skill-authoring and Anthropic's official docs

The original's core concepts (trigger conditions, quality gates, automatic/explicit modes, retrospective) are preserved unchanged.

Installation

Individual repo (recommended)

**User-level** (available in all projects):

# macOS / Linux
git clone https://github.com/abhattacherjee/claudeception.git ~/.claude/skills/claudeception

# Windows
git clone https://github.com/abhattacherjee/claudeception.git %USERPROFILE%\.claude\skills\claudeception

**Project-level** (available only i