israeli-property-appraisal

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md’s Step 2a/2b ingests outsider-authored free text via the user-provided address (and the manual “paste the rows back” route), which is then used to call Govmap/Nadlan endpoints and process returned deal rows.

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

• Secret detected (high risk: 1.00). I found a high-entropy, literal API token in the bundled script. The line API_TOKEN = "cf153c72-fb28-4b27-9db4-982bc89cb3b0" is a real-looking UUID-like key (not a placeholder like YOUR_API_KEY). It is used in requests to govmap/nadlan endpoints (see code comments and usage), so it is a usable client token embedded in the repository. Although the comment says it is the public client constant embedded in the nadlan.gov.il front end, that does not make it a harmless placeholder — it is a literal credential in code and should be treated as a secret (or at least rotated/managed appropriately). All other strings inspected are documentation, configuration names, numeric policy figures, or low-entropy examples and were ignored per the rules.