command-aso-pass-play
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow contract that accepts external input, creating a potential attack surface for indirect prompt injection.
- Ingestion points: The workflow contract in
SKILL.mdspecifies anargument-hint: "<context>", which indicates that the agent will ingest and process untrusted data provided in the execution context. - Boundary markers: There are no explicit delimiters, XML-style tags, or instructions provided to the agent to treat the ingested context as untrusted or to ignore embedded instructions within that data.
- Capability inventory: No specific subprocess calls, file-system write operations, or network requests were identified within the analyzed files (
SKILL.md,agents/openai.yaml). However, the skill mentions dispatching tools and calling related skills which are not present in this scope. - Sanitization: No evidence of sanitization, validation, or escaping of the input context was found.
Audit Metadata