recoup-internal-consulting-objection-handler
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process 'raw objections' provided by prospects or users. This data enters the agent's context to be classified and personalized, which presents a surface for malicious instructions embedded within the objection text to influence the agent's behavior.
- Ingestion points: The workflow takes a 'raw objection' as the primary input for the consulting process (SKILL.md).
- Boundary markers: There are no explicit instructions or delimiters defined to isolate the user-provided objection from the agent's system instructions or to ignore embedded commands within that input.
- Capability inventory: The skill has the capability to read workspace-specific files (e.g., AGENTS.md, DESIGN.md), access internal reference files, and invoke sibling skills like 'recoup-internal-consulting-tasteful-design'.
- Sanitization: The instructions do not specify any validation, filtering, or sanitization steps for the input objection before it is processed by the personalization logic.
- [SAFE]: The skill correctly scopes its operations to the current workspace and authorized entity folders. It uses a vendor-specific configuration file ('brand.lock.json') and refers to a sibling capability ('recoup-internal-consulting-tasteful-design') which are consistent with the 'recoupable' vendor identity.
Audit Metadata