Chat History Analyzer
v1.0.5Extracts and analyzes Cursor IDE chat history to identify key discoveries, obstacles, and solutions, saving findings to the journal.
⭐ 1· 428·5 current·5 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description claim to extract and analyze Cursor IDE chat history; the packaged Python scripts explicitly open Cursor's local SQLite files under the user's home directory, parse messages, analyze them with pattern matching, and save markdown reports to the OpenClaw journal directory — this is coherent and proportional to the stated purpose.
Instruction Scope
SKILL.md and the scripts instruct the agent to read Cursor's local storage (global and workspace state.vscdb), parse recent messages, and write reports to ~/.openclaw/journal. These actions are within scope but involve reading potentially sensitive chat content from a user's machine; SKILL.md includes example hard-coded paths (e.g., /Users/ghost/...) that are just examples but could confuse users if copied verbatim. There are no instructions that read unrelated system areas or send data to external endpoints.
Install Mechanism
There is no external install/download step (instruction-only skill with embedded Python scripts). No network downloads, package installs, or archive extraction are used. The code relies on Python and sqlite3 which are typical on target systems.
Credentials
The skill declares no required environment variables, credentials, or config paths beyond standard user-home locations. It does import a sibling 'self-optimizer' module (via sys.path manipulation) — that dependency is reasonable for combined log analysis but means the overall behavior depends on the code of that other skill.
Persistence & Privilege
always is false and the skill is user-invocable; SKILL.md suggests running it as an hourly cron job (expected for periodic analysis). Autonomous agent invocation is allowed by default (not flagged here), so if the agent is permitted to schedule or run this skill it will read local chat data on each run — this is a privacy consideration rather than a code-level privilege escalation.
Assessment
This skill appears to do what it says: reading Cursor's local SQLite chat storage, scanning recent messages for discoveries/obstacles/solutions, and writing reports to your OpenClaw journal. Before installing, consider the following:
- Privacy: the skill reads local chat history files that may contain sensitive secrets or private content. Only install if you trust the skill and want those chats processed and stored in ~/.openclaw/journal.
- Review code: the repository includes Python scripts; inspect them (and the referenced self-optimizer skill) if you have privacy or security concerns. The skill imports a sibling 'self-optimizer' module via sys.path — that dependency influences runtime behavior.
- Cron examples: SKILL.md contains example absolute paths with a specific username (e.g., /Users/ghost/...); adjust paths to your environment rather than copying them verbatim.
- Limit execution scope: if you plan to schedule it, consider running manually first or restricting the cron to run under a user account with appropriate access. Monitor the journal directory for files the skill creates.
- If you need higher assurance: run the scripts in a sandbox or examine runtime logs to confirm no outbound network activity occurs. If you cannot review the code, treat the skill as sensitive because it accesses private chat data.Like a lobster shell, security has layers — review code before you run it.
latestvk979zdbwnmgjve9z6zhk2kd6k982hyyc
License
MIT-0
Free to use, modify, and redistribute. No attribution required.