storyclaw-autoposter
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill transmits the user's StoryClaw API key and social media content to an external API endpoint (https://aipuejhjwmabtobjrqdz.supabase.co). This is the intended functional behavior required for the skill to interact with the StoryClaw service.
- [COMMAND_EXECUTION]: The skill instructs the agent to use a
curlcommand to upload local media files to the service's backend. This operation is used as a prerequisite when a user provides local images or videos for posting. - [PROMPT_INJECTION]: The skill ingests and processes user-supplied text for social media 'optimization' and publishing, which creates a surface for indirect prompt injection. This risk is mitigated by a mandatory confirmation flow in SKILL.md and SKILL.zh-CN.md that requires the user to review and explicitly confirm the final content and target platforms before any API calls are made.
Audit Metadata