graphify-review

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the graphify command-line utility with review-delta and review-analysis subcommands. These operations are performed on local repository data to identify the 'blast radius' and test gaps of staged changes, which aligns with the skill's stated purpose.
  • [PROMPT_INJECTION]: The skill identifies an attack surface for indirect prompt injection by processing untrusted data from the repository.
  • Ingestion points: Staged or modified files in the local repository are read for analysis (SKILL.md).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded content are defined in the instructions.
  • Capability inventory: Subprocess execution via the graphify CLI tool (SKILL.md).
  • Sanitization: No sanitization or validation of the repository file content is specified before the analysis is performed.
  • Note: This finding represents a standard operating surface for code analysis tools and does not exhibit malicious intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 03:40 AM
Security Audit — agent-trust-hub — graphify-review