acceptance-contract
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, untrusted data from various sources such as project specifications, tickets, and existing contracts.\n
- Ingestion points: The process begins by resolving and reading source materials, existing contracts, and pending amendments as described in Step 1 of the contract building phase in SKILL.md.\n
- Boundary markers: The instructions do not specify any delimiters or explicit 'ignore instructions' warnings when handling the source content, which may lead the agent to interpret embedded commands in those documents as valid instructions.\n
- Capability inventory: The skill itself is purely instructional and does not include scripts, network operations, or file-system write capabilities; however, the agent following these instructions might utilize other tools in its environment.\n
- Sanitization: There are no mentioned mechanisms for sanitizing or validating the content extracted from external sources.
Audit Metadata