debugging-wizard

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill's design involves ingesting potentially untrusted data such as error logs, stack traces, and code snippets while maintaining access to system tools (e.g., Python debugger, Node inspector, and Git). This combination creates an attack surface where malicious instructions could be embedded in the analyzed content to manipulate the agent's behavior during debugging.
  • Ingestion points: The agent is instructed to gather error messages, stack traces, and log entries as documented in SKILL.md and references/systematic-debugging.md.
  • Boundary markers: The skill does not provide specific instructions for using delimiters or ignoring embedded commands within the ingested data.
  • Capability inventory: The skill facilitates the use of python, node, git, and dlv via the shell to investigate issues.
  • Sanitization: There are no mentioned processes for sanitizing or validating the content of external logs or code before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 05:45 AM
Security Audit — agent-trust-hub — debugging-wizard