spec-executor
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it is designed to ingest and act upon instructions contained within external data sources.
- Ingestion points: The agent is instructed to find the latest
SPEC READYblock in inherited conversation history and read referenced spec artifacts (inSKILL.md). - Boundary markers: The instructions do not specify the use of delimiters or provide guidance to distinguish between data and instructions within the ingested specifications, increasing the risk of the agent obeying malicious commands hidden in the spec.
- Capability inventory: The skill allows for local code implementation, testing via
/tdd, and potentially sensitive external actions such as code pushes, reviews, or deployments if authorized by the input specification (inSKILL.md). - Sanitization: The skill lacks input sanitization mechanisms to filter or escape instructions in the specs, though it does mandate the redaction of credentials in the final output receipt.
Audit Metadata