skills/vinta/hal-9000/write-like-me/Gen Agent Trust Hub

write-like-me

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to rewrite or draft English prose based on user-provided input, such as READMEs, blog posts, emails, and PR comments. This ingestion of untrusted data creates a surface for indirect prompt injection if the source text contains malicious instructions intended to hijack the agent's behavior.
  • Ingestion points: SKILL.md (Workflow Step 2: "gather the user's draft from the conversation or the files they pointed at").
  • Boundary markers: Absent. The instructions do not specify the use of delimiters or provide warnings to ignore instructions embedded within the processed text.
  • Capability inventory: The skill has the ability to output text to the conversation and "edit files in place" (SKILL.md Workflow Step 5).
  • Sanitization: Absent. No instructions are provided to sanitize, escape, or validate the input text before the agent processes it for stylistic rewriting.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 07:48 AM
Security Audit — agent-trust-hub — write-like-me