ascii-diagram-validator

Pass

Audited by Gen Agent Trust Hub on Oct 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through its 'Self-Evolving Skill' feature, which instructs the agent to persist changes to the skill's instructions based on observed behavior during the processing of external, untrusted documentation files. * Ingestion points: The skill reads and processes user-controlled markdown documentation files (e.g., docs/*.md) for ASCII diagram validation. * Boundary markers: The instructions do not specify any delimiters or safety warnings to treat the content of the processed files as untrusted data. * Capability inventory: The skill uses Bash to execute its validation script and is authorized by its own instructions to update its SKILL.md file if workarounds or fixes are needed. * Sanitization: No sanitization or verification of the content within the code blocks of analyzed markdown files is performed before the agent might use that context to improve the skill's instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 25, 2026, 10:08 AM