test-against-spec
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it instructs the agent to derive expected behavior from potentially untrusted external sources such as user requests, issues, and product documentation. This data could contain hidden instructions intended to subvert the agent's testing or code-fixing logic.
- Ingestion points: Specification sources (user request, issue, mini-spec, product documentation, ADR) as defined in the 'Boundaries' and 'Workflow' sections of
SKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or explicit 'ignore instructions' warnings when the agent processes these external documents.
- Capability inventory: The agent is granted the capability to execute actions through real interfaces (API, CLI, library entry points) and to modify source code when the
fixqualifier is used, as detailed inSKILL.mdandreferences/fix-mode.md. - Sanitization: There are no instructions for sanitizing, validating, or filtering the content of the external specification sources before they are used to drive agent actions.
- [NO_CODE]: The analyzed skill consists solely of Markdown documentation and instructions. It does not package any executable scripts, binaries, or configuration files that would trigger automated dependency installation or code execution during the skill's deployment.
Audit Metadata