Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Security: added --sandbox to all pandoc export calls (prevents Lua filter/LaTeX RCE).
What This Skill Does
Agent Chronicle generates AI-written diary entries from an agent's perspective after each session. Entries run 400-600 words and include sections for wins, frustrations, learnings, emotional state, and notable interactions. Optional modules track memorable quotes, open questions, decision reasoning, and relationship patterns over time.
Unlike session logs that capture outputs and errors, this records the agent's subjective experience including emotional tone and evolving patterns, making it useful for continuity across sessions.
When to use it
- Generating daily reflective entries after coding sessions
- Collecting memorable quotes from human collaborators
- Reviewing past decisions and their reasoning
- Tracking mood and topic patterns across weeks
- Exporting diary archives to PDF or HTML for monthly review
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Write today's diary entry
- 1Runs generate.py --today --emit-task to produce a task JSON with session context and prompts
- 2Spawns a sub-agent via sessions_spawn using the system and user prompts from the task JSON
- 3Sub-agent writes a 400-600 word reflective markdown entry
- 4Pipes the output into generate.py --today --from-stdin to save the entry to memory/diary/
A saved markdown file at memory/diary/YYYY-MM-DD.md with sections for summary, projects, wins, frustrations, learnings, and emotional state
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.