execute-plan
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external implementation plans and repository content to drive its execution logic, creating a vulnerability surface where instructions embedded in data could influence agent behavior.
- Ingestion points: The agent is instructed to read implementation plans (including HTML format), repository instructions, and perform repository inspection as the primary inputs for its tasks (SKILL.md).
- Boundary markers: The instructions lack explicit directives to treat the content of implementation plans as untrusted data or to ignore instructions embedded within plan metadata or HTML structure.
- Capability inventory: The skill possesses high-privilege capabilities including repository modification (implementing code), invocation of sub-agents (subagent-dispatch), and execution of verification tests.
- Sanitization: No sanitization, escaping, or validation logic is specified for the plan content before it is used to determine execution steps.
Audit Metadata