Self Improving Cognition
Security checks across malware telemetry and agentic risk
Overview
This is a coherent self-improvement skill, with the main consideration being that it stores personal cognition and decision notes locally and includes a small shell helper plus a declared suite dependency.
This skill appears safe for its stated purpose, but treat the generated cognition logs as personal data. Install it only if you are comfortable keeping those records in your workspace, and review the bundled helper script and declared self-improving-agent dependency if you plan to use the broader suite.
VirusTotal
65/65 vendors flagged this skill as clean.
Risk analysis
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.
Your private reflections about thinking patterns, stress, habits, and decisions may be saved for later use by the agent.
The skill tells the agent to maintain persistent local cognition records that may include sensitive self-assessment, lifestyle, and stress information.
Append to `.learnings/cognition/BASELINE.md` ... Environmental & Lifestyle Factors ... Sleep ... Nutrition ... Exercise ... Stress
Use it only in a workspace where you are comfortable storing personal self-assessment notes, and periodically review or delete the .learnings/cognition files if they are no longer needed.
Running the helper script will execute local shell commands and create an exercise log file in the current directory.
The package includes an interactive shell script that, if run, prompts the user and writes a local Markdown exercise record. The script is simple and purpose-aligned, with no network or destructive behavior shown.
#!/bin/bash ... read -p "请选择(1-4): " exercise_type ... cat > "cognitive-exercise-$timestamp.md"
Run the script only intentionally and from a directory where you want the generated exercise record to be saved.
Installing or using this skill may involve related suite behavior depending on how the platform handles declared dependencies.
The skill declares a dependency on another self-improvement skill while the registry summary lists no install spec. This appears suite-related, but users should notice the dependency relationship.
"dependencies": ["self-improving-agent"], "partOfSuite": "self-improving-suite"
Review the referenced self-improving-agent dependency before relying on the full suite.