菜单
Use when the user wants to review a Claude Code session for quality — analyzes the current session (or a specified transcript path) for prompting effectiveness, agent performance, and environment gaps, producing actionable recommendations
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when analyzing a large corpus of text, code, or data that exceeds a single agent's effective context - orchestrates parallel Worker subagents, Critic review subagents, and a final Summarizer subagent with task tracking and failure recovery
Use after initial design context is gathered, before brainstorming - resolves contradictions in requirements, disambiguates terminology, clarifies scope boundaries, and verifies assumptions to prevent building the wrong solution
Use when creating or developing anything, before writing code or implementation plans - refines rough ideas into fully-formed designs through structured Socratic questioning, alternative exploration, and incremental validation
Use when executing implementation plans with independent tasks in the current session - dispatches fresh subagent for each task, reviews once per phase, loads phases just-in-time to minimize context usage
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
Use when completing tasks, implementing major features, or before merging to verify work meets requirements - dispatches code-reviewer subagent, handles retries and timeouts, manages review-fix loop until zero issues
| name | review-session |
| description | Use when the user wants to review a Claude Code session for quality — analyzes the current session (or a specified transcript path) for prompting effectiveness, agent performance, and environment gaps, producing actionable recommendations |
Review a single Claude Code session transcript for quality and produce actionable findings.
Do not use nested subagents. This workflow may dispatch the conversation-reviewer agent, but that agent must analyze the transcript directly and must not dispatch additional subagents.
ed3d-extending-claude plugin must be installed.ed3d-session-reflection plugin must be installed (provides the conversation-reviewer agent and reduce-transcript.py script).The user may invoke this as:
/review-session — review the current session/review-session /path/to/transcript.jsonl — review a specific transcriptIf no argument is provided, use the current session's transcript path (from the SessionStart hook context injection).
If an argument was provided, use it as the transcript path. Otherwise, use the current session's transcript path from the SessionStart hook injection.
If you cannot determine the transcript path, tell the user:
I don't know the current session's transcript path.
Either provide a path: /review-session /path/to/session.jsonl
Or ensure the ed3d-session-reflection SessionStart hook is active.
Run the reduction script to produce a token-efficient version:
python3 "${CLAUDE_PLUGIN_ROOT}/scripts/reduce-transcript.py" "<transcript_path>" "/tmp/session-review/reduced.txt"
Tell the user you're reducing the transcript for analysis.
Create a temporary output path for findings:
/tmp/session-review/findings.md
Dispatch the conversation-reviewer agent with:
Transcript path: /tmp/session-review/reduced.txt Write your findings to: /tmp/session-review/findings.md
Read the transcript, analyze it, and write your findings following your output format. Do not dispatch or invoke any subagents.
Once the reviewer completes, read /tmp/session-review/findings.md and present the findings to the user.
Present the full findings — do not summarize or truncate. The reviewer has already calibrated the length to be proportional to what's interesting.