muapi-blog-header
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [DATA_EXFILTRATION]: The skill transmits data and an authentication token ($MUAPI_API_KEY) to the external domain api.muapi.ai using curl commands.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the muapi CLI tool for authentication and image generation tasks.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided content via the topic input and interpolates it directly into the prompt used for image generation.
- Ingestion points: The
topicinput field in SKILL.md. - Boundary markers: No delimiters, escaping, or explicit instructions are used to distinguish user input from the rest of the generation prompt.
- Capability inventory: The skill possesses network access via
curland CLI execution capabilities via themuapitool. - Sanitization: There is no evidence of filtering or validation performed on the
topictext before it is used in prompt construction. - [EXTERNAL_DOWNLOADS]: The skill relies on external infrastructure and a CLI tool (muapi) that are not part of the standard platform environment.
Audit Metadata