recoup-internal-consulting-integration-onboarding
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from
AGENTS.mdandintegrations/AGENTS.mdfiles within the workspace. While it provides manual boundary markers by instructing the agent to exclude unrelated or personal material, there is no automated sanitization for these inputs. The skill's capabilities, including scaffolding code and executing network reads, present a surface for indirect prompt injection. - [DYNAMIC_EXECUTION]: The workflow facilitates the creation of executable code by using a sibling skill and the
recoup-internal-consulting-integration-scaffoldertool to generate adapter source code from templates at runtime. - [SAFE]: The skill demonstrates positive security posture by explicitly directing the agent and user to keep credentials, secret values, and raw data caches out of documented files and tracked source control.
Audit Metadata