Contact
Productivity Failure patterns

Retro

LLM-driven session retrospection. Detects friction in agent sessions (mechanical pre-pass + LLM inference + cross-session) and materializes approved learnings into user memory, project rules, skill PRs, checkpoints, or harness artifacts. Commands: /retro, /retro outcome, /retro audit.

Latest release v3.2.0

Install

Choose a supported route below. Package contents, agent registration and runtime dependencies can differ between installers.

  1. Claude Code via this marketplace 
    /plugin install retro@netresearch-claude-code-marketplace

    First add the marketplace once in Claude Code: /plugin marketplace add netresearch/claude-code-marketplace. Then install the selected plugin. For setup and installation scopes, see Claude Code plugin setup A plugin can include several skills and additional components.

  2. npx interactive skill selection 
    npx skills add https://github.com/netresearch/retro-skill

    Requires Node.js 22.20+. Run from the target project, then choose the source skills and agents in the installer. Uses the skills CLI from skills.sh Installs skill instructions. Extra plugin hooks, commands and external runtime tools are not installed by this route.

  3. composer require PHP project, as a package 
    composer require netresearch/retro-skill

    Allow the Composer skill installer to activate and approve skill registration when prompted. Requires netresearch/composer-agent-skill-plugin Discovers trusted skill packages and registers their instructions in AGENTS.md. Runtime tools may need separate setup.

  4. composer skills:add PHP project, direct source 
    composer skills:add netresearch/retro-skill

    Install and activate netresearch/composer-agent-skill-plugin first; approve the selected source's skill registration. Provided by netresearch/composer-agent-skill-plugin Pins source repositories without a Packagist package and records them in composer.skills.lock. Runtime dependencies remain separate.

What you can do with it

  • LLM-driven session retrospection
  • Detects friction in agent sessions (mechanical pre-pass + LLM inference + cross-session) and materializes approved learnings into user memory, project rules, skill PRs, checkpoints, or harness artifacts
  • Commands: /retro, /retro outcome, /retro audit

What you get back

  • Actionable retrospective proposals grouped by their intended destination; durable changes are applied only after each proposal is approved.

What it expects in your project

  • Claude Code: The host that runs the plugin and the /retro command
  • uv: Runs every Python script on the Python 3.11 or later its header declares (PEP 723), downloading that interpreter when the host has none; python3 alone is not enough, macOS ships 3.9. For mechanical polling detection, session scope and review feedback (detect-mechanical.py, derive-session-scope.py, collect-review-findings.py) it also installs the shell parser tree-sitter-bash the scripts declare; the first run downloads it
  • jq: Skill discovery, manifest parsing, and the optional hook
  • gh and/or glab: Creating pull/merge requests for skill updates
  • git: Cloning source repos and materializing changes

Tags

  • productivity
  • skill-skills