← Back to Skills Marketplace
5468
Downloads
5
Stars
12
Active Installs
17
Versions
Install in OpenClaw
/install agent-chronicle
Description
AI-powered diary generation for agents - creates rich, reflective journal entries (400-600 words) with Quote Hall of Fame, Curiosity Backlog, Decision Archae...
Usage Guidance
Review the privacy tradeoff before installing. Use it only if you want local session history turned into persistent diary material. Disable auto-generation, memory integration, quote capture, and relationship tracking unless you explicitly want them, avoid storing secrets or regulated data in session logs, and prefer PDF or local-styled exports over HTML that loads remote CSS.
Capability Analysis
Type: OpenClaw Skill
Name: agent-chronicle
Version: 0.7.2
Agent Chronicle is a comprehensive journaling and self-reflection tool for OpenClaw agents. The skill gathers context from session logs and persistent memory files to generate reflective diary entries, weekly digests, and mood analytics. It follows OpenClaw best practices by using the platform-native `sessions_spawn` pattern for sub-agent generation and respects workspace boundaries via environment variables. While the scripts perform file operations and use `subprocess` for document exports (via `pandoc` and `weasyprint`), these actions are strictly aligned with the stated features, and no evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
The scripts and documentation match the stated purpose: generating agent diary entries, digests, exports, mood analytics, quote tracking, curiosity tracking, decision logs, and optional relationship notes from local workspace memory.
Instruction Scope
The trigger metadata includes broad ordinary terms such as diary, journal, quotes, curious, decisions, mood, and digest; for a skill that reads prior session context and writes persistent reflections, accidental activation is a material concern.
Install Mechanism
No install-time execution, package scripts, dependency install hooks, obfuscation, or hidden setup behavior were found; setup is an explicit Python onboarding script.
Credentials
Workspace memory access is purpose-aligned: the skill reads local memory logs and diary files, truncates context, uses environment variables for workspace detection, and does not directly call external LLM APIs. HTML export can reference a remote CSS CDN.
Persistence & Privilege
The skill intentionally creates long-lived local files under memory/diary and can append summaries to daily memory logs, extract quote/curiosity/decision/relationship sections into separate files, resurface old entries, and produce exports, with no clear retention, deletion, or per-item review controls.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install agent-chronicle - After installation, invoke the skill by name or use
/agent-chronicle - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.7.2
agent-chronicle v0.7.2
- Updated documentation for clarity and current workflow, including changing examples and references from /tmp to $TMP_DIR.
- Minor edits to SKILL.md for improved usability and onboarding.
- Version bumps in metadata and package files.
- No functional changes to core skill logic.
v0.7.1
agent-chronicle v0.7.1
- Adds mood analytics, weekly digests, and "On This Day" resurfacing features.
- Introduces cron-based auto-generation for daily entries.
- New dedicated templates for curiosity, daily, decisions, quotes, relationships, and weekly sections.
- Configuration and packaging improvements with `config.example.json`, `package.json`, and lockfile.
- Updated documentation and onboarding for new features and automation options.
v0.6.6
Docs portability cleanup: replaced hardcoded temp file examples with portable TMP_DIR placeholders.
v0.7.0
- added 2, updated 6 file(s).
- Updated SKILL.md and bundle contents.
v0.6.5
fix: config.json now uses relative diary_path (not absolute /root/...), memory_integration disabled by default — addresses ClawHub security review
v0.6.4
v0.6.4: sync README/CHANGELOG across ClawHub, GitHub and local; version badges updated
v0.6.3
Synced changelog/docs, aligned package metadata, added .clawignore
v0.6.2
Security: added --sandbox to all pandoc export calls (prevents Lua filter/LaTeX RCE).
v0.6.1
Fix: SKILL.md metadata updated to openclaw format. Added .gitignore.
v0.6.0
v0.6.0: Refactor to OpenClaw-native sessions_spawn pattern. Remove raw HTTP calls, use platform defaults, add CONTRIBUTORS.md. Thanks to @lejean2000 for architectural feedback.
v0.5.0
v0.5.0: Beautiful PDF export (Velvet Edition) + OpenClaw Gateway integration + Privacy cleanup
v0.4.1
v0.4.1: Model flexibility - removed Haiku requirement, skill now works with any capable model. Recommends Claude models for best results.
v0.4.0
v0.4.0: AI-powered diary generation with Claude Haiku, includes Quote Hall of Fame, Curiosity Backlog, Decision Archaeology, and Relationship Evolution
v0.3.2
Add ClawHub runtime requirements metadata (SKILL.md frontmatter)
v0.3.1
Add runtime requirements metadata for ClawHub discovery
v0.3.0
- Setup now runs automatically on first use of generate commands; manual setup remains available.
- Setup wizard includes new option to integrate summaries with daily memory logs.
- Updated documentation for a streamlined, quicker onboarding experience.
- Minor documentation improvements and clarifications in SKILL.md.
- Updated example configuration and script usage to match setup changes.
- No breaking changes to existing daily entry format or commands.
v0.2.0
Agent Chronicle 0.2.0 introduces structured, AI-perspective daily journaling.
- Adds rich daily entry templates capturing projects, wins, frustrations, learnings, emotional state, quotes, curiosities, decisions, and relationship notes.
- Fully documents philosophy, triggers, onboarding, commands, and configuration.
- Includes export tools (PDF, HTML), mood/topic analysis, and specialized backlogs (quotes, curiosities, decisions, relationship).
- Enables both auto-generated and interactive entry writing, plus convenient reading and summary commands.
- Enhances self-reflection and continuity for agent experiences.
Metadata
Frequently Asked Questions
What is Agent Chronicle?
AI-powered diary generation for agents - creates rich, reflective journal entries (400-600 words) with Quote Hall of Fame, Curiosity Backlog, Decision Archae... It is an AI Agent Skill for Claude Code / OpenClaw, with 5468 downloads so far.
How do I install Agent Chronicle?
Run "/install agent-chronicle" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Agent Chronicle free?
Yes, Agent Chronicle is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Agent Chronicle support?
Agent Chronicle is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Agent Chronicle?
It is built and maintained by Robby (@robbyczgw-cla); the current version is v0.7.2.
More Skills