track-unrest-events

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt embeds a literal-looking API key value (X-WorldMonitor-Key: wm_0123456789abcdef0123456789abcdef01234567) and instructs using that header, which encourages including secret values verbatim in generated requests or code.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow calls a public RPC that returns outsider-authored unrest feed fields (e.g., titles/summaries/actors from ACLED/GDELT/RSS) as readable JSON, which the agent will place into the LLM context for analysis.

HIGH W008: Secret detected in skill content (API keys, tokens, passwords).

• Secret detected (high risk: 1.00). The file contains a high-entropy literal that looks like an API key: "wm_0123456789abcdef0123456789abcdef01234567" on line [10]. It is not an obvious placeholder (e.g., YOUR_API_KEY or sk-xxxx) and matches the form of a service API key (prefix "wm_" + long hex-like string). Per the secret definition, this should be treated as a potential real credential and flagged. Other values (e.g., "Authorization: Bearer ...") are generic placeholders and are not flagged.