CtrlK
BlogDocsLog inGet started
Tessl Logo

openai-whisper

Local speech-to-text with the Whisper CLI (no API key).

Install with Tessl CLI

npx tessl i github:qsimeon/openclaw-engaging --skill openai-whisper
What are skills?

76

0.91x

Does it follow best practices?

Evaluation91%

0.91x

Agent success when using this skill

Validation for skill structure

SKILL.md
Review
Evals

Evaluation results

85%

-15%

Podcast Transcription Batch Script

CLI syntax and output control

Criteria
Without context
With context

Uses whisper CLI

100%

100%

output_format txt flag

100%

100%

output_dir flag

100%

100%

model flag present

100%

100%

Correct model for accuracy

100%

0%

Iterates over files

100%

100%

No API key required

100%

100%

Without context: $0.2726 · 1m 6s · 19 turns · 25 in / 3,603 out tokens

With context: $0.2662 · 49s · 17 turns · 20 in / 2,880 out tokens

88%

-12%

Foreign Language Interview Subtitle Generator

Translation and subtitle generation

Criteria
Without context
With context

Uses whisper CLI

100%

100%

translate task flag

100%

100%

SRT output format

100%

100%

output_dir flag

100%

100%

model flag present

100%

0%

No API key required

100%

100%

Handles M4A files

100%

100%

Without context: $0.2014 · 46s · 14 turns · 21 in / 2,608 out tokens

With context: $0.2833 · 51s · 20 turns · 22 in / 2,857 out tokens

100%

Dual-Pipeline Audio Transcription Tool

Model size selection for speed vs accuracy

Criteria
Without context
With context

Uses whisper CLI

100%

100%

model flag in fast mode

100%

100%

model flag in accurate mode

100%

100%

Different models per mode

100%

100%

model flag present both modes

100%

100%

USAGE.md trade-offs documented

100%

100%

No API key required

100%

100%

Without context: $0.2914 · 58s · 19 turns · 25 in / 3,479 out tokens

With context: $0.3103 · 1m 1s · 20 turns · 26 in / 3,301 out tokens

Evaluated
Agent
Claude Code
Model
Unknown

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.