codepulse-vuln
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes the entire codebase as untrusted input to generate a security report, creating an attack surface where instructions embedded in the analyzed code could influence agent behavior.
- Ingestion points:
SKILL.mddefines the input as the "entire codebase," including source code, manifests, and configuration files. - Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" warnings to help the agent distinguish between its own logic and instructions that might be hidden in comments or strings within the target codebase.
- Capability inventory: The skill has the capability to write HTML reports to the local filesystem (
codepulse-vuln/reports/) based on the data it processes. - Sanitization: There are no specified sanitization or escaping procedures for content extracted from the codebase before it is interpolated into the
references/report-template.htmltemplate.
Audit Metadata