debugging-protocol
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest and analyze external, untrusted data such as application logs and API payloads. This creates a surface for indirect prompt injection. However, the protocol mitigates this risk through a structured workflow that uses formal templates and specific validation tasks.\n
- Ingestion points:
SKILL.md(Step 5: Execution and Documentation),languages/frontend.md(Data & Rendering analysis).\n - Boundary markers: Debugging findings are recorded in a dedicated template (
assets/debugging-session-template.md).\n - Capability inventory: Interaction with various system-level diagnostic tools (e.g.,
jstack,gdb,pprof) and file-writing for session documentation.\n - Sanitization: No explicit sanitization of log content is mandated.\n- [EXTERNAL_DOWNLOADS]: The language-specific guides suggest using and installing standard, well-known diagnostic utilities and libraries from official package repositories (npm, PyPI, RubyGems). These downloads are characteristic of the software development use case and are sourced from trusted well-known services.
Audit Metadata