arete-deploy
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data files that could contain malicious instructions designed to influence the agent's behavior during deployment.
- Ingestion points: The skill reads
idl/program.jsonandartifacts/program.program-spec.json(references/program-publication.md), as well as.arete/<Stack>.stack-manifest.json(SKILL.md). - Boundary markers: Absent; the instructions do not specify the use of strict delimiters or specific 'ignore embedded instructions' prompts for the content of these files.
- Capability inventory: The skill can execute network-mutating commands via the vendor CLI, including
a4 program push,a4 up, anda4 stack delete(SKILL.md, references/deployment-lifecycle.md). - Sanitization: The skill includes instructions to validate the file's kind, program ID, and artifact hash before proceeding (references/program-publication.md).
Audit Metadata