Pattern Recognition
v1.0.0Identifies, learns, and applies patterns from operations, errors, and resources to generate templates, analyze efficiency, and suggest optimizations.
⭐ 0· 54·1 current·1 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 match the included Python scripts: both learner and analyzer operate on operation history, logs, and metrics. One minor inconsistency: SKILL.md shows CLI usage (pattern ...) but no install spec or CLI wrapper is provided in the package — the repo contains Python scripts but not an installed 'pattern' command. Functionality requested (reading workspace markdown/log/metrics) is aligned with the stated purpose.
Instruction Scope
Runtime instructions and the scripts read files under /home/openclaw/.openclaw/workspace (memory/*.md, logs/*.log, metrics/*.json) and write patterns/suggestions/logs back into a patterns directory. This is appropriate for pattern-learning but means the skill will process potentially sensitive agent memory and logs; there are no network calls or external endpoints in the code.
Install Mechanism
No install spec (instruction-only) and only two Python scripts are included. This is low-risk from an install standpoint because nothing is downloaded or extracted from external URLs.
Credentials
The skill requests no environment variables or external credentials. It hardcodes a workspace path (/home/openclaw/.openclaw/workspace) which is reasonable for an agent-local analysis tool but gives it access to whatever lives in that workspace.
Persistence & Privilege
always is false and the skill does not modify other skills or global agent settings. It persists learned state and logs under the workspace patterns directory only.
Assessment
This skill appears to do what it says: it scans the agent workspace for operations, logs, and metrics and writes learned patterns and suggestions back to a patterns folder. Before installing, review the contents of /home/openclaw/.openclaw/workspace (memory, logs, metrics) to ensure no sensitive secrets or credentials would be processed or exposed. Note there is no network/exfiltration code in the included scripts, but the SKILL.md references a 'pattern' CLI that isn't provided — verify how the skill will be invoked in your environment (you may need to run the Python scripts directly or provide a wrapper). If you want a safer test, run the scripts in a sandboxed environment with sample data first.Like a lobster shell, security has layers — review code before you run it.
latestvk978dwcjpx8emjpjkd9pb32fts8434xj
License
MIT-0
Free to use, modify, and redistribute. No attribution required.