codebase-onboarding
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its core function of reading and processing untrusted data from an external repository.\n
- Ingestion points: The skill reads various manifest files (e.g.,
package.json,pom.xml,pyproject.toml,pubspec.yaml), source code files (.java,.ts,.py,.dart), and configuration files (application.yml) through shell commands and file read tools.\n - Boundary markers: The skill does not instruct the agent to use specific delimiters or to disregard embedded instructions when processing content from these files, increasing the risk of the agent following instructions found within the analyzed data.\n
- Capability inventory: The skill has access to
BashandWritetools, which provides a functional path for an injection to result in unauthorized file system modifications or command execution.\n - Sanitization: The skill lacks logic to sanitize, escape, or filter the content of the files it reads before presenting that data to the LLM for analysis.
Audit Metadata