reviewing-user-research-prior-art-survey
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The agent is instructed in
SKILL.mdto run a local script,validate_user_research_prior_art.py, with parameters derived from artifact file paths. This represents an execution surface for local commands as part of the validation process. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and evaluate untrusted external artifacts, such as research vocabulary maps and search outputs. This data could contain malicious instructions designed to subvert the agent's review logic.
- Ingestion points: Artifacts are provided as input for review in the
SKILL.mdworkflow. - Boundary markers: The instructions do not provide explicit boundary markers or directives to ignore instructions embedded within the research artifacts being analyzed.
- Capability inventory: The agent can execute local scripts and perform network re-probes of external sources as defined in
references/conditions.md(Condition C13). - Sanitization: No content sanitization or instruction filtering is performed on the ingested research data.
Audit Metadata