Elevenlabs

Converts text to natural speech using ElevenLabs for clinical and healthcare use cases. Use when generating patient instructions, discharge summaries, medica...

Security Scan

The skill's requirements and instructions align with its stated purpose (ElevenLabs TTS/Music for clinical use); it only needs an ElevenLabs API key and otherwise contains no code or unusual installation steps.

Purpose & Capability

Name/description (ElevenLabs TTS & music for clinical scenarios) match the declared requirement (ELEVENLABS_API_KEY) and the SKILL.md/README guidance. Nothing requested (no unrelated env vars, binaries, or config paths) appears out of scope.

Instruction Scope

SKILL.md instructs the agent to call ElevenLabs HTTP APIs or to use an existing TTS tool configured for ElevenLabs. This is appropriate for the stated goal, and the docs explicitly warn not to include PHI in logs or external calls. Note: because the skill may cause content to be sent to a third-party API, operators must ensure clinical data governance (e.g., do not send protected health information unless allowed by policy/BAA with ElevenLabs).

Install Mechanism

Instruction-only skill (no install spec, no code files to execute). This is the lowest-risk install model: nothing is downloaded or written by the skill itself.

Credentials

Only ELEVENLABS_API_KEY is required and listed as primaryEnv, which is proportionate. The README/SKILL.md suggests storing the key either in env or in ~/.openclaw/openclaw.json; storing secrets directly in a plaintext config file is convenient but can be insecure — prefer a secret manager or environment/vault integration in production.

Persistence & Privilege

always:false and no install scripts or system changes are requested. The skill does not request persistent elevated privileges or modify other skills' configs. Agent-autonomous invocation is enabled by default but that is normal for skills and not a concern by itself here.

Assessment

This skill appears coherent and limited to calling ElevenLabs using your ELEVENLABS_API_KEY. Before enabling it: (1) Decide which API key to use (prefer an org-owned key and monitor billing/usage). (2) Do not send protected health information (PHI) to third-party APIs unless your organization has an appropriate legal/compliance agreement with ElevenLabs (e.g., BAA) and has approved the data flow. (3) Store the key securely (environment variable or secret vault) rather than hardcoding it into plaintext config files. (4) Monitor API usage and logs for accidental data leaks and set quotas if possible.

ElevenLabs Text-to-Speech for Clinical Projects

Quick-start skill for OpenClaw Clinical Hackathon participants. Use ElevenLabs TTS for patient-facing voice: instructions, reminders, discharge info, and accessible content.

Prerequisites

• ELEVENLABS_API_KEY — Set in environment or in ~/.openclaw/openclaw.json under skills.entries.elevenlabs.apiKey (or env.ELEVENLABS_API_KEY).
• Get a key at ElevenLabs (free tier available).

When to Use This Skill

• Patient discharge or aftercare instructions (spoken).
• Medication or appointment reminders.
• Multilingual health messages (e.g. 30+ languages).
• Accessibility: turning written clinical text into clear speech.
• Long-form content (e.g. patient education) with natural, empathetic tone.

How to Use

1. Ensure the TTS tool is available
   If your OpenClaw setup has a text-to-speech tool (e.g. tts_text_to_speech or similar), use it with ElevenLabs as the provider. The tool will use ELEVENLABS_API_KEY when configured for ElevenLabs.

2. When the user asks for spoken output

   • Prefer short, clear sentences for instructions and reminders.
   • For medical terms, use ElevenLabs’ pronunciation controls or a pronunciation dictionary if available to improve accuracy.
   • Suggest a calm, professional voice for clinical content.

3. If calling the API directly

   • Endpoint: POST https://api.elevenlabs.io/v1/text-to-speech/{voice_id}
   • Headers: xi-api-key: <ELEVENLABS_API_KEY>, Content-Type: application/json
   • Body: {"text": "<content>", "model_id": "eleven_multilingual_v2"} (or eleven_flash_v2_5 for low latency).
   • Prefer eleven_multilingual_v2 for non-English or mixed-language clinical text.

Best Practices for Clinical TTS

• Tone: Use a warm, clear, professional voice; avoid overly casual or dramatic tones.
• Chunking: Break long text into short paragraphs or bullets to improve clarity and pacing.
• Language: Set the correct language (or use a multilingual model) for the patient’s preferred language.
• Sensitive content: Do not include PHI or other sensitive data in log messages or external calls; keep API usage consistent with your security and compliance setup.

Quick Reference

Use case	Suggestion
Short reminders	eleven_flash_v2_5 for speed
Long-form / multilingual	eleven_multilingual_v2
Medical terminology	Use pronunciation hints or dictionary if supported

Getting Help

• ElevenLabs Text-to-Speech API
• ElevenLabs Healthcare use cases