de-ai-flavor

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted text from external documents to perform rewrites and edits.
  • Ingestion points: The skill utilizes the 'Read', 'Grep', and 'Glob' tools to ingest data from user-provided file paths (e.g., 'docs/*.md') and git diff outputs.
  • Boundary markers: There are no specific delimiters or 'ignore embedded instructions' markers defined in 'SKILL.md' to isolate the text being processed.
  • Capability inventory: The skill has the 'Edit' capability, which allows it to modify the local filesystem based on potentially malicious instructions embedded in the ingested content.
  • Sanitization: No sanitization, validation, or filtering logic is specified for the input text before the agent is instructed to 'Rewrite' or 'Simplify' it.
  • [NO_CODE]: The skill does not contain any executable scripts, binary files, or remote dependencies, relying entirely on the agent's internal reasoning and standard filesystem tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 06:41 AM