Stats
Actions
Tags
From learning-agents
Investigates identified issues in LearningAgent sessions by analyzing transcripts, determining root causes like knowledge gaps or missing docs, and updating YAML issue files with reports.
How this skill is triggered — by the user, by Claude, or both
Slash command
/learning-agents:investigate-issuesThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Research identified issues from a LearningAgent session to determine their root causes.
Research identified issues from a LearningAgent session to determine their root causes.
$ARGUMENTS is the path to the session log folder (e.g., .deepwork/tmp/agent_sessions/<session_id>/<agent_id>/).
Session log folder structure:
!cat learning_agents/doc/learning_log_folder_structure.md 2>/dev/null
Agent used: !cat $ARGUMENTS/agent_used 2>/dev/null || echo "unknown"
Identified issues to investigate:
!grep -l 'status: identified' $ARGUMENTS/*.issue.yml 2>/dev/null || echo "(none)"
Additional investigation guidelines:
!learning_agents/scripts/cat_agent_guideline.sh $ARGUMENTS issue_investigation
-------- CURRENT KNOWLEDGE OF AGENT --------
!learning_agents/scripts/generate_agent_instructions_for_session.sh $ARGUMENTS
------ END CURRENT KNOWLEDGE OF AGENT-------
If no identified issues are listed above, report that and stop.
Refer back to the conversation_transcript.jsonl file as needed in this process.
For each issue file with status identified:
issue_description or locate lines near seen_at_timestampscore-knowledge.mdFor each investigated issue, use Edit to update the issue file:
status: identified to status: investigatedinvestigation_report field:status: investigated
investigation_report: |
<Root cause analysis with specific transcript line numbers as evidence.
Explain what knowledge gap or instruction deficiency caused the issue.>
Simply say "Session log folder done."
issue_description — only add the investigation_reportidentified issuenpx claudepluginhub unsupervisedcom/deepwork --plugin learning-agentsAnalyzes LearningAgent session transcripts to identify mistakes, knowledge gaps, underperformance, and creates YAML issue files for problems found.
Analyzes Claude Code Orchestrator/Worker/Reviewer transcripts for given issues to detect structural problems, protocol deviations, and anti-patterns.
Analyzes Claude Code JSONL transcripts to detect anti-patterns, tool misuse, user frustration signals, and workflow patterns using DuckDB SQL, 10 dimensions, and PM4Py mining.