runtime-logger

Log agent activities to .claude/logs/activity.log with standard log levels.

19|1|Updated Nov 25, 2025
One-click install
npx skills add https://github.com/Dowwie/tasker --skill runtime-logger
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: runtime-logger
Source: https://github.com/Dowwie/tasker/tree/main/internal/skills/runtime-logger
Command: npx skills add https://github.com/Dowwie/tasker --skill runtime-logger

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Logs agent activities to a file during task execution, providing a traceable record of actions and decisions.

Core Features & Use Cases

  • Creates and uses a persistent log at .claude/logs/activity.log to capture lifecycle events (start, tool usage, decisions, completion).
  • Supports standard log levels (INFO, DEBUG, WARN, ERROR) to categorize events and facilitate troubleshooting.
  • Facilitates auditing and debugging of multi-step workflows and subagent activity in deterministic pipelines.

Quick Start

Create the log directory and start emitting logs to .claude/logs/activity.log during task execution.

Frequently Asked Questions about runtime-logger

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

FAQPage Schema
How do I log agent activities to a file for task traceability?

Agent activities are logged to a file during task execution at .claude/logs/activity.log, providing a persistent traceable record of actions and decisions for debugging and auditing.

What log levels are supported for debugging agent workflows?

Standard log levels including INFO, DEBUG, WARN, and ERROR are supported to categorize agent lifecycle events and facilitate troubleshooting of deterministic task workflows.

Does this logging approach work with subagent orchestration pipelines?

Auditing and debugging of multi-step workflows and subagent activity in deterministic pipelines is supported, capturing lifecycle events like start, tool usage, decisions, and completion.

How do I set up the log directory before writing activity logs?

The logging process ensures the log directory exists before use, automatically creating the .claude/logs directory structure to start emitting logs to activity.log.

What is the best way to audit decisions in automated task workflows?

Persisting key activity logs to a file during task execution provides a traceable record of decisions, enabling auditing of multi-step workflows and subagent activity in deterministic pipelines.

Are there limitations when using file-based logging for subagent activity?

Logs are written to a single persistent file at .claude/logs/activity.log, which may constrain traceability when concurrent subagents emit interleaved events in complex orchestration scenarios.