code-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute
git remote get-url originandgit log -1 --format='%h %s'to retrieve repository metadata. These commands are used to populate a metadata table in the generated analysis document and are standard for development workflows. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted source code, which creates an attack surface for indirect prompt injection via malicious instructions embedded in code comments or strings.
- Ingestion points: Source code files read during the analysis workflow (SKILL.md).
- Boundary markers: The skill does not specify the use of delimiters or instructions to ignore embedded commands within the analyzed code.
- Capability inventory: The agent can read local files and execute shell commands (git).
- Sanitization: There is no explicit sanitization or filtering of the code content before it is processed or included in the output document.
Audit Metadata