typefully
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from external social media platforms via the Typefully API, creating a vector where adversarial instructions could be embedded in post drafts, replies, or analytics data.
- Ingestion points: Untrusted content enters the agent's context through several commands in
scripts/typefully.js, includingdrafts:list,drafts:get,analytics:posts:list, andqueue:get. - Boundary markers: While the skill provides high-level safety guidelines (e.g., 'Golden Rules' in
references/posting-workflow.md), it does not implement technical delimiters or specific instructions for the agent to ignore prompts contained within the social media data itself. - Capability inventory: The skill has permissions to perform network operations (API calls to
api.typefully.com) and read/write access to the local file system for post content and configuration management. - Sanitization: The script outputs raw content retrieved from the API to the console; there is no evidence of filtering or escaping logic to neutralize potential natural language instructions within the social media text.
Audit Metadata