elevenlabs-tts
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates an attack surface where untrusted data from external files is ingested and processed.
- Ingestion points: The
textargument inscripts/generate_audio.pyis intended to be populated by external content, as shown inSKILL.mdexamples using command substitution. - Boundary markers: There are no delimiters or instructions provided to the agent to ignore or isolate potential commands embedded within the input text.
- Capability inventory: The skill uses the
requestslibrary for network egress to ElevenLabs and possesses file system write access to save audio files. - Sanitization: No sanitization, escaping, or validation is applied to the input text before it is transmitted to the external API.
- [COMMAND_EXECUTION]: The skill's documentation explicitly guides the agent to perform shell command execution, including command substitution and piping (e.g.,
$(cat video_script.txt)andsed), which increases the risk of unintended command execution if filenames or file contents are manipulated.
Audit Metadata