The Agent Skills Directory

[SAFE]: The skill performs its primary function of voice cloning using the vendor's infrastructure. No malicious or suspicious patterns were found during the analysis.\n- [COMMAND_EXECUTION]: The skill executes a local script, scripts/voice_clone_api.py, which coordinates interactions with the Giggle.pro API using the Python requests library.\n- [CREDENTIALS_UNSAFE]: The skill requires a GIGGLE_API_KEY. The implementation follows security best practices by instructing the agent to utilize system environment variables, preventing the exposure of the key in shell history or process listings.\n- [EXTERNAL_DOWNLOADS]: The skill makes network requests to https://giggle.pro to submit audio samples and retrieve synthesized results. This communication is restricted to the official vendor domain and is required for the skill's operation.\n- [PROMPT_INJECTION]: The skill provides an attack surface for indirect prompt injection by accepting user-supplied text and audio URLs.\n
Ingestion points: User input for audio_url, voice_id, and text is passed to the execution script in SKILL.md.\n
Boundary markers: None present; the skill does not use delimiters to wrap untrusted user input in the script execution.\n
Capability inventory: Access to network operations via the requests library and subprocess execution via the exec tool.\n
Sanitization: Input is parsed by the Python argparse module, which provides standard command-line argument handling and validation for basic types.

giggle-voice-clone