cc-skill-continuous-learning
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of a configuration file and a shell script that performs basic session analysis. It reads from a transcript path provided by the environment and stores results in a local directory (
~/.claude/skills/learned/). No network communication, remote script execution, or sensitive data exfiltration was observed. - [PROMPT_INJECTION]: The skill involves the agent reviewing its own session transcripts. While this creates a surface for indirect prompt injection if a transcript contains malicious text, the script itself is a benign monitoring tool that does not parse or execute the content of the messages.
- Ingestion points:
evaluate-session.shaccesses the transcript viaCLAUDE_TRANSCRIPT_PATHto count user messages. - Boundary markers: None.
- Capability inventory: Local file system access for directory creation and standard text processing tools (
grep,sed,jq). - Sanitization: Not applicable to the script's function of counting messages.
Audit Metadata