From log-agent-issue
Logs structured issues (errors, warnings, info) from subagents to JSONL file via fire-and-forget Bash script with bun. Use for non-blocking tracking in multi-agent Claude Code workflows.
How this skill is triggered — by the user, by Claude, or both
Slash command
/log-agent-issue:log-agent-issueThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Structured issue logger for subagents. Appends JSONL entries to `~/.claude/logs/agent-issues.jsonl`.
Structured issue logger for subagents. Appends JSONL entries to ~/.claude/logs/agent-issues.jsonl.
Log issues when you encounter:
Pipe a JSON payload to the script via Bash with run_in_background: true:
echo '{"agentId":"<your-agent-id>","issues":[{"type":"error","message":"<what happened>","toolName":"<tool>","context":{"key":"value"},"suggestion":"<how to fix>"}]}' | bun ${CLAUDE_PLUGIN_ROOT}/src/log-issue.ts "${CLAUDE_SESSION_ID}" 2>/dev/null
{
"agentId": "triage-worker",
"issues": [
{
"type": "error | warning | info",
"message": "Human-readable description",
"toolName": "optional — MCP tool that failed",
"filePath": "optional — file involved",
"context": "optional — object with extra details",
"suggestion": "optional — remediation hint"
}
]
}
run_in_background: true — never block on logging2>/dev/null — silent failure, never interrupt workflowissues array, don't make multiple callstimestamp and sessionId automaticallynpx claudepluginhub nathanvale/side-quest-marketplace-old --plugin log-agent-issueCreates kebab-case YAML issue files tracking problems observed in LearningAgent sessions from log folder paths and descriptions. Useful for real-time failure reporting via direct invocation or identify skill.
Files structured GitHub bug reports for agent-validator by collecting runtime evidence (config, debug logs, execution state) and drafting issues. Useful when users encounter a suspected defect.
Guides adding try-catch console logging for successes, errors, and state to make app behavior visible to AI agents in debugging and agentic coding workflows.