continuous-learning
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves its intended purpose of knowledge management and implements quality gates to ensure information is verified and safe for reuse.
- [COMMAND_EXECUTION]: The skill uses file system tools to create and manage markdown files within the .claude/skills/ directory as part of its core functionality.
- [EXTERNAL_DOWNLOADS]: The skill uses web research tools to supplement learned information with current best practices and official documentation. These actions are bounded by the skill's instructions to cite sources and verify information.
- [PROMPT_INJECTION]: The skill manages an attack surface for indirect prompt injection by processing session data and codebase content to generate new instructions. This is evaluated as safe due to built-in mitigations.
- Ingestion points: Session history and codebase files read during debugging (SKILL.md).
- Boundary markers: Uses user confirmation prompts as a verification gate before finalizing any new skill content.
- Capability inventory: Accesses file writing, skill registration, and network search tools.
- Sanitization: Includes explicit directives to exclude sensitive data such as credentials, API keys, and internal URLs from generated skills.
Audit Metadata