social-post-writer-seo
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill takes user-supplied inputs such as 'Topic' and 'Keyword' to generate social media posts, which presents a surface for indirect prompt injection. However, the risk is negligible as the skill lacks capabilities to perform file operations, network requests, or command execution.
- Ingestion points: Template variables including
{Insert Topic},{Insert Keyword},{Target Audience},{Tone}, and{Brand}in SKILL.md. - Boundary markers: Present; the prompt uses structured Markdown headers (e.g., ### INPUT, ### GOAL) to separate untrusted user data from instructions.
- Capability inventory: None; the skill is limited to text generation and does not utilize any external tools or system APIs.
- Sanitization: None; the template relies on the model's inherent safety filters and formatting instructions.
- [NO_CODE]: The skill consists entirely of Markdown documentation and a prompt template. There are no scripts, binaries, or configuration files that execute logic on the underlying system.
Audit Metadata