api-caller

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs users (and therefore the agent) to pass API keys and Bearer tokens directly via command-line --headers (and shows an example with "Bearer TOKEN"), which requires embedding secret values verbatim in generated commands — a high exfiltration risk.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). In scripts/parse_openapi.py, the required workflow fetches an outsider-provided OpenAPI spec URL at runtime (fetch_spec via urlopen) and parses its operation fields, meaning attacker-authored free text in the spec is ingested without selecting a specific item first.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches external OpenAPI specs at runtime (e.g., https://petstore3.swagger.io/api/v3/openapi.json, https://api.fakevendor.com/v2/openapi.json, https://api.weather.gov/openapi.json) and parses them to drive which operations the agent discovers and executes, meaning remote content can directly control the agent's behavior.