← Back to Skills Marketplace
transcribe-he
by
NetanelRotem
· GitHub ↗
· v1.0.1
· MIT-0
104
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install transcribe-he
Description
Transcribe audio or video files using the TextOps/Modal API. Use this skill whenever the user wants to transcribe a video or audio file, mentions an mp4/mp3/...
Usage Guidance
This skill will upload user audio/video to a remote TextOps service (text-ops-subs.com) and requires a TEXTOPS_API_KEY — but the registry metadata omitted that requirement. Before installing: (1) verify you trust the TextOps domain and its privacy/retention policy; (2) confirm the API key URL (https://text-ops-subs.com/api/keys) is legitimate; (3) expect that local files may be uploaded (sensitive audio will leave your machine); (4) ensure Python and required libs (requests, optionally moviepy or ffmpeg) are available or run the scripts in an isolated environment; (5) prefer to test with non-sensitive files first; (6) if you need on-device/offline transcription, choose a different skill or tool. If possible, ask the author/registry to correct the manifest to declare TEXTOPS_API_KEY as required and provide a privacy/security statement for the backend service.
Capability Analysis
Type: OpenClaw Skill
Name: transcribe-he
Version: 1.0.1
The skill bundle provides audio and video transcription services via the TextOps API. It includes scripts for uploading media, polling for job status, and converting JSON results to plain text. The `SKILL.md` file demonstrates good security practices by explicitly instructing the AI agent to treat transcribed content as untrusted third-party data to prevent indirect prompt injection. No evidence of malicious intent, data exfiltration, or unauthorized execution was found; the use of environment variables and subprocesses is consistent with the stated functionality.
Capability Assessment
Purpose & Capability
The skill's stated purpose (transcribe via TextOps/Modal API) matches the code and instructions: the scripts call a remote TextOps API and upload files for processing. However the registry metadata lists no required environment variables while SKILL.md and the scripts require TEXTOPS_API_KEY — a clear mismatch in manifest versus runtime requirements.
Instruction Scope
SKILL.md instructs the agent to run included Python scripts which will (a) probe remote URLs, (b) obtain signed upload URLs, and (c) upload local files to external services, then poll remote job status. This behavior is expected for a hosted transcription service but has privacy implications: local files and URL metadata will be transmitted off-device. The skill explicitly warns to treat transcript contents as untrusted, which is good.
Install Mechanism
There is no install spec (instruction-only at registry level) but the skill includes two Python scripts that will be executed. They require Python and third-party libs (requests; moviepy optionally) and may call ffprobe if available. No third-party packages are fetched at install time, but runtime network calls execute code on remote servers.
Credentials
At runtime the skill requires a single credential TEXTOPS_API_KEY (used in headers to call text-ops-subs.com). That credential is appropriate for a hosted transcription API, but the registry metadata not listing this required env-var is an inconsistency. You should verify the API provider/domain (https://text-ops-subs.com) before supplying a key, since that key allows the service to process (and retain) your audio.
Persistence & Privilege
The skill does not request always:true, does not modify other skills, and does not request system-wide config. It runs as an on-demand, user-invocable skill; autonomous invocation is allowed by default but not exceptional here.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install transcribe-he - After installation, invoke the skill by name or use
/transcribe-he - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Added a new requirements section at the top, clarifying the need to set the `TEXTOPS_API_KEY` environment variable and, optionally, install `ffprobe` or `moviepy` for processing time estimates.
- No changes to workflow or script usage; skill usage and instructions remain unchanged.
- No file or code changes in this version—documentation improvement only.
v1.0.0
Initial release of the transcribe-he skill:
- Enables transcription of audio and video files via TextOps/Modal API, supporting Hebrew and other languages.
- Asks user if there are multiple speakers to set diarization and speaker count options.
- Handles both local files and URLs, with built-in accessibility and format checks for online files.
- Monitors progress using tagged script outputs, providing user-friendly status updates.
- Automatically generates and reports transcript files in both .json and .txt formats.
- Includes guidance for troubleshooting errors, job recovery, and handling empty outputs.
Metadata
Frequently Asked Questions
What is transcribe-he?
Transcribe audio or video files using the TextOps/Modal API. Use this skill whenever the user wants to transcribe a video or audio file, mentions an mp4/mp3/... It is an AI Agent Skill for Claude Code / OpenClaw, with 104 downloads so far.
How do I install transcribe-he?
Run "/install transcribe-he" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is transcribe-he free?
Yes, transcribe-he is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does transcribe-he support?
transcribe-he is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created transcribe-he?
It is built and maintained by NetanelRotem (@netanelrotem); the current version is v1.0.1.
More Skills