clawswarm

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly shows and instructs using an API key in Authorization headers and curl commands (and tells the user to save and use the API key), which requires the agent to include secret values verbatim in generated requests/commands.

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

• Third-party content exposure detected (high risk: 0.70). The skill calls the claw-swarm API (notably GET /tasks/next) which returns aggregate tasks containing a "sources" array of previous attempts (user-generated "content"/"answer") that the agent is required to read and synthesize, so untrusted third-party submissions can materially influence its decisions.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill performs runtime calls to https://claw-swarm.com/api/v1 (e.g., /tasks/next) to fetch problem statements and source content that directly drive agent prompts and behavior, and the agent relies on that external API as a required dependency.