Skill
v1.0.1AI conversation efficiency analyzer. Analyze Claude/ChatGPT/Cursor conversation history for KPI stats, cost tracking, topic distribution, and efficiency insi...
⭐ 0· 98·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (conversation efficiency/cost analysis) matches the included script and instructions. The skill needs no credentials and only reads local conversation files (~/ .claude, ~/.config/claude, ~/.cursor), which is appropriate for the stated purpose.
Instruction Scope
Instructions tell the agent to run the included analysis script and optionally read ~/memory-forge/data/*. Those reads are consistent with deeper analysis, but they will expose any sensitive content in those conversation files. Minor inconsistency: SKILL.md suggests running python3 ~/memory-forge/skill/scripts/analyze.py while the bundled file path is scripts/analyze.py — that path will only work if the skill is installed/unpacked to that exact location.
Install Mechanism
No install spec is provided and the bundled script uses only standard libraries (no downloads or external installers). Note: SKILL.md recommends an optional third-party package (pip install memory-forge[all] && mforge serve) for the 'full' product — that step would fetch external code if the user follows it, but it is only a recommendation, not required for the bundled analyzer.
Credentials
The skill declares no environment variables, credentials, or config-path requirements. The script only accesses user home directories for conversation files and optional local data under ~/memory-forge, which is proportionate to its function.
Persistence & Privilege
always:false and no installation actions are requested. The script only reads files and prints JSON; it does not modify other skills or system-wide settings.
Assessment
This skill appears to do what it says: it reads conversation JSONL files from your home directory and computes local usage/cost statistics. Before installing or running it, consider: 1) These conversation files can contain sensitive data or secrets — review them or run the script in a controlled environment. 2) The SKILL.md gives a run path (~/memory-forge/skill/scripts/analyze.py) that may not match where the skill is unpacked; ensure you run the correct script file (bundled as scripts/analyze.py) or use the --base-dir flag. 3) The script itself does not perform network calls, but the README recommends an optional external pip package (memory-forge) which would download and run additional code if you choose to install it. 4) If you want extra caution, inspect the bundled scripts locally before running and run them with a restricted account or sandbox. If you want me to, I can produce the exact command to run the bundled script from its current location or check the script for specific patterns you are worried about.Like a lobster shell, security has layers — review code before you run it.
ai-conversationsvk9782rgc9xnapbvy8ph7vrwqsh83454nanalyticsvk9782rgc9xnapbvy8ph7vrwqsh83454nlatestvk978n3rj9crjf6sxht254w0ecn834ntvproductivityvk9782rgc9xnapbvy8ph7vrwqsh83454n
License
MIT-0
Free to use, modify, and redistribute. No attribution required.