teach-impeccable
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from the local codebase to influence the agent's persistent configuration.
- Ingestion points: The skill explicitly instructs the agent to scan various project files, including READMEs, documentation,
package.json, and source code components to discover design patterns. - Boundary markers: There are no defined delimiters or instructions provided to the agent to ignore or isolate potentially malicious instructions embedded within the scanned codebase files.
- Capability inventory: The skill utilizes file-read operations to gather data and a file-write operation to create or update
CLAUDE.mdin the project root, which serves as a persistent instruction set for future agent sessions. - Sanitization: The instructions do not include steps to sanitize, escape, or validate the content gathered from the codebase before it is synthesized into the final design context configuration.
Audit Metadata