What does the 'Verdict' in the report mean?

The Verdict provides a PASS or FAIL status based on whether the actual session logs align with the mandatory steps and hooks defined in your skill documentation.

Can this skill detect subagent failures?

Yes, it extracts SubagentStart and SubagentStop events from debug logs to confirm if agents like 'reviewer' or 'gap-analyzer' were called in the correct order.

Where are Claude Code session logs stored?

The skill automatically searches for logs in ~/.claude/projects/ and detailed execution traces in ~/.claude/debug/.

How do I start a session analysis?

To analyze a session, provide the specific session UUID and the file path to the target SKILL.md you wish to validate against.

Does it check if temporary files were deleted?

Yes, the tool verifies artifact lifecycle by searching for file creation events and confirming subsequent deletion commands or filesystem states.

Session Analyzer for Claude Code

Name: Session Analyzer for Claude Code
Author: team-attention

byteam-attention

•

313

•

Analytics & Monitoring

Analyzes completed Claude Code sessions to verify tool execution, hook triggers, and artifact creation against specification files.

Session Analyzer is a specialized debugging and verification tool designed for developers building complex workflows in Claude Code. It automatically retrieves session logs and debug traces, compares actual execution history against the requirements defined in a SKILL.md file, and generates a comprehensive report. This is particularly useful for identifying why a subagent failed to trigger, verifying that security hooks were correctly applied, or ensuring that temporary artifacts were properly cleaned up after a task completion, ensuring high-reliability AI agent deployments.

Key Features

01313 GitHub stars

02Comparison of actual behavior vs. SKILL.md specifications

03Comprehensive pass/fail reporting with deviation summaries

04Detailed tracking of subagent invocations and hook events

05Automated session log and debug trace retrieval

06Artifact verification for file creation and deletion

Use Cases

01Debugging complex multi-agent workflows that failed unexpectedly

02Verifying compliance with defined development patterns and artifact management

03Validating that security hooks and tool matchers are triggering correctly

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add team-attention/plugins-for-claude-natives session-analyzer

For use in Claude.ai and ChatGPT

Download Skill