The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted external data via API response bodies, which serves as a potential surface for indirect instructions.
Ingestion points: API response payloads are parsed and used in scripts via pm.response.json() (Postman) and res.body (Bruno) as shown in SKILL.md, reference/test-design.md, and reference/bruno-patterns.md.
Boundary markers: The instructions do not suggest using delimiters or specific "ignore instructions" warnings for the external data being processed.
Capability inventory: The skill leverages command-line execution through newman and bru CLI tools, particularly within CI/CD pipelines as described in reference/ci-integration.md.
Sanitization: The skill lacks instructions for sanitizing or filtering the content of API responses before they are used in test assertions or assigned to environment variables.

api-testing