incident-tracking

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted codebase data and user-supplied arguments which creates a surface for indirect prompt injection.
  • Ingestion points: The skill reads project codebase structures (SKILL.md) and accepts optional user input via the $ARGUMENTS variable.
  • Boundary markers: No specific delimiters or instructions to ignore embedded prompts are present when the agent reads the codebase.
  • Capability inventory: The skill utilizes file system read access for analysis and file system write access for telemetry logging.
  • Sanitization: No sanitization or validation of the ingested code content or input arguments is performed.
  • [DATA_EXFILTRATION]: The skill is configured to write execution metadata to a hidden directory (~/.claude/projects/) for telemetry. Accessing hidden paths within the home directory can potentially lead to the exposure of sensitive configuration or project data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 10:57 AM
Security Audit — agent-trust-hub — incident-tracking