competitor-benchmark

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt includes a config snippet that instructs adding an "Authorization: Bearer YOUR_API_KEY" header (i.e., embedding an API key directly in agent configuration), which requires handling a secret value verbatim and risks it being emitted or exfiltrated.

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow calls Promptwatch MCP tools (e.g., get_competitor_heatmap, get_citations) that will ingest competitor/mention/citation text sourced from third-party web content and/or other users’ data via the MCP server, which is outsider-authored free text entering the agent’s LLM context through those tool results.