PavelGuzenfeld

Debrief Agent — Development skill for Claude Code

Development community

Post-session review that proposes changes to the agent's environment — never to the code.

How to install Debrief Agent

Installs to ~/.claude/skills/pavelguzenfeld-agent-sdlc-debrief-agent/SKILL.md

Terminal

mkdir -p ~/.claude/skills/pavelguzenfeld-agent-sdlc-debrief-agent && curl -fsSL https://raw.githubusercontent.com/PavelGuzenfeld/agent-sdlc/HEAD/commands/debrief-agent.md -o ~/.claude/skills/pavelguzenfeld-agent-sdlc-debrief-agent/SKILL.md

Restart Claude Code, or start a new session, for it to be picked up.

What Debrief Agent does

description: Post-session review that proposes changes to the agent's environment — never to the code. Nine lenses over the transcript, ranked findings, filed through /done's gate on request.

Debrief agent

Review one session's transcript and propose changes to the environment it ran in: rules, hooks, skills, memory, permissions. Never propose a code change — that is `/complicate` or `/simplify`'s job. Edits nothing itself; filing is the only write, and only on request.

Target

Alternatives in Development

Rank — /rank - Triage Scraped Jobs into a Ranked Shortlist 36.6k ★
Breach Check — HIBP k-anonymity check on a password wordlist 4.5k ★
Fidelity — Measure how faithfully a clone reproduces a site — pixel-diff plus motion-fidelity into one 0-100 score, a let 3.6k ★