x-post-writer
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user inputs such as topics, angles, and rough drafts to generate social media content.
- Ingestion points:
SKILL.md(Step 1) collects user-supplied topics and draft text to inform the writing process. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for the interpolation of these user inputs.
- Capability inventory: The skill possesses the capability to communicate with external platforms via
lib.publish(SKILL.mdStep 8) and generate images vialib.illustrate(SKILL.mdOptional illustration). - Sanitization: The skill includes a "Humanizer pass" and an "Optional audit" (
SKILL.mdSteps 5-6) which scrub AI-like vocabulary markers, though these are stylistic rather than security-focused. The requirement for a human to review the "Approval card" (SKILL.mdStep 7) before publishing is the primary safeguard.
Audit Metadata