Agent Chronicle
ReviewAudited by ClawScan on May 1, 2026.
Overview
Agent Chronicle is a coherent journaling skill, but users should understand it persists private session-derived diary content and can optionally run automated generation.
Before installing, review `config.json` for `diary_path`, `memory_integration`, privacy level, and `auto_generate`; disable memory or automation features you do not want, and avoid recording secrets in diary entries.
Findings (5)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Private conversations, quotes, decisions, and relationship notes may become long-term agent memory.
The skill can persist diary-derived summaries into the agent's broader memory, which is aligned with journaling but may retain private interaction details for future use.
Automatically add diary summaries to your main daily memory log (`memory/YYYY-MM-DD.md`). ... Enabled by default during setup.
Review `config.json` before use, disable `memory_integration` if unwanted, and avoid recording secrets or sensitive personal data.
Session or diary context used for reflection may be included in sub-agent prompts.
Diary generation is designed to pass generated context and prompts to an OpenClaw sub-agent, which is purpose-aligned but means diary/session context is shared with another model run.
spawn a **sub-agent** via `sessions_spawn` ... system: `task.system` ... user: `task.prompt`
Use this only with trusted OpenClaw model/provider settings and review generated task content when handling sensitive sessions.
If connected to a scheduler, diary entries may be generated without a separate prompt each time.
The shipped configuration includes auto-generation enabled, and the documentation describes cron auto-generation; no hidden scheduler installation is shown, but users should be aware of recurring non-interactive generation if they configure cron.
"auto_generate": true
Set `auto_generate` to `false` unless you intentionally want scheduled diary creation, and review any cron/OpenClaw automation separately.
Running export uses local conversion tools on diary content and writes output files.
Export commands invoke the local `pandoc` binary to convert diary files, which is expected for PDF/HTML export and uses argument lists plus pandoc sandboxing.
subprocess.run(["pandoc", str(temp_md), "-o", str(output_path), "--sandbox"
Only use export on trusted diary content, keep local conversion tools updated, and verify the output path.
Users who want the optional PDF exporter may need to install extra packages themselves.
The optional PDF exporter depends on manually installed Python packages that are not automatically installed by this skill.
WeasyPrint is required. Install with: pip3 install weasyprint ... Python-Markdown is required. Install with: pip3 install markdown
Install optional packages from trusted sources and pin versions if using this in a controlled environment.