End-of-session adversarial review loop. Assemble the session's work into a role-assigned, self-contained brief, then run independent reviewers in parallel — an isolated code-reader (the idea-validator agent) that reads the ACTUAL files and web-checks technology currency, plus an external-family model if you have one — synthesize where they agree vs diverge, apply the cheap-safe fixes…
Install
npx skills add awrshift/agent-memory-kit --skill session-reviewnpx skills add awrshift/agent-memory-kit --skill session-review -a claude-code -gnpx skills add awrshift/agent-memory-kit --skill session-review -a codex -ggemini skills install https://github.com/awrshift/agent-memory-kit.git --path plugins/memory-kit/skills/session-reviewgh skill install awrshift/agent-memory-kit session-reviewgit clone --depth 1 https://github.com/awrshift/agent-memory-kit /tmp/awrshift__agent-memory-kit
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/awrshift__agent-memory-kit/plugins/memory-kit/skills/session-review ~/.claude/skills/session-review # Claude Code
cp -R /tmp/awrshift__agent-memory-kit/plugins/memory-kit/skills/session-review ~/.agents/skills/session-review # Codex, Cursor, Gemini CLI, Copilot, OpenCode