detecting-debug-endpoints
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes text content from external web server responses. A malicious server could return content designed to hijack the agent's instructions.\n
- Ingestion points: The
scripts/probe_debug.pyscript reads the body of HTTP responses from user-specified targets.\n - Boundary markers: Findings are reported through structured
lib.reportobjects, separating target data from the agent's instructional logic.\n - Capability inventory: The skill utilizes network read capabilities (
safe_get) and local script execution.\n - Sanitization: The script applies regular expression fingerprints to ensure the response content matches expected technical patterns before reporting a finding.
Audit Metadata