skills.rest
AMDResearch
AMDResearchOFFICIAL
Explore all skills by this author →
GitHub

session-review

Parse session capture files and generate structured Session Review artifacts.

3|3|Updated Aug 19, 2024
Search Tags:#retrospective#handoff#improvement#process#lessons#session-review
One-click install
npx skills add https://github.com/AMDResearch/dh_comms --skill session-review-amdresearch
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: session-review
Source: https://github.com/AMDResearch/dh_comms/tree/main/.agents/skills/session-review
Command: npx skills add https://github.com/AMDResearch/dh_comms --skill session-review-amdresearch

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Review session captures and extract actionable improvements to shorten cycle times, improve outcomes, and reduce repeated issues.

Core Features & Use Cases

  • Parses a session capture to identify goals, actions, blockers, and outcomes.
  • Produces a Session Review artifact detailing what went well, what failed, and concrete next steps (local proposals or forward-to-meta lessons).
  • Integrates with failure patterns and improvement workflows stored under .agents to maintain a learning system.

Quick Start

Place a session capture at .untracked/session-captures/<timestamp>-<agent>.md and run the session-review skill to generate an improvement report.

Frequently Asked Questions about session-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn agent session captures into actionable lessons and process improvements?▼

To turn session captures into actionable lessons, you parse the capture file to identify goals, actions, blockers, and outcomes, then generate a formal improvement report. This produces concrete next steps like local proposals or forward-to-meta lessons.

What is a session review and when do I need to analyze agent failure modes?▼

A session review is a structured artifact detailing what went well, what failed, and concrete next steps. You need this analysis when you want to identify failure modes and positive patterns from session captures to shorten cycle times and reduce repeated issues.

How do I generate an improvement report from a session capture file?▼

Place your session capture at .untracked/session-captures/<timestamp>-<agent>.md and run the session-review skill. It orchestrates failure-mode identification and positive patterns, outputting a structured review to .agents/improvement/session-reviews.

Can I trace local proposals and forward-to-meta lessons from a retrospective?▼

Yes, the retrospective process outputs structured reviews to .agents/improvement/session-reviews and, when warranted, routes findings to local-proposals or forward-to-meta folders. This maintains traceability and sharing for process improvement workflows.

Do I need a specific file format to review session blockers and outcomes?▼

You need a markdown capture file located at .untracked/session-captures/<timestamp>-<agent>.md. The review process parses this specific file structure to identify goals, actions, blockers, and outcomes for the improvement artifact.

What's the best way to maintain a learning system for repeated agent issues?▼

The best way is to integrate failure patterns and improvement workflows stored under .agents. By consistently reviewing session captures and extracting forward-to-meta lessons, you maintain a continuous learning system that reduces repeated issues.

Related Skills

Explore More Content & Communication Skills→
AMDResearchAMDResearch
3

session-capture

Generate a normalized session capture file with metadata, goals, actions, and decisions.

Official
Advanced
AMDResearchAMDResearch
3

lessons-forward

Forward generalizable template findings to the Agentic Meta Project.

Official
Intermediate
agenticsorgagenticsorg
25

ReasoningBank Intelligence

Implement adaptive learning and meta-cognitive capabilities for AI agents using ReasoningBank.

Official
Advanced
clovernguyen1010-ship-itclovernguyen1010-ship-it

bmad-retrospective

Coordinate structured Agile epic retrospectives to extract lessons and capture action items.

Community
Advanced