openai-whisper-api

Transcribe audio via OpenAI Audio Transcriptions API (Whisper).

2.27x

Quality

70%

Does it follow best practices?

Impact

91%

2.27x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./openclaw/skills/openai-whisper-api/SKILL.md

Evaluation results

100%

65%

Podcast Transcription Automation

Speaker-hinted transcription with custom output path

Criteria

Without context

With context

Uses transcribe.sh

100%

No direct curl call

100%

Speaker prompt flag

100%

Custom output path flag

100%

Output extension correct

50%

100%

Script accepts audio file argument

50%

100%

API key reference

100%

Model not incorrectly overridden

100%

transcripts/ directory created

100%

Without context: $0.1331 · 38s · 11 turns · 16 in / 2,030 out tokens

With context: $0.3534 · 1m 17s · 20 turns · 278 in / 4,422 out tokens

80%

33%

Field Interview Transcription Pipeline

Language-hinted batch transcription

Criteria

Without context

With context

Uses transcribe.sh

100%

No direct curl call

100%

Language flag used

50%

100%

Loops over .ogg files

100%

Default output naming

46%

API key documented

100%

Directory argument

100%

Model not incorrectly overridden

100%

openclaw config mentioned

Without context: $0.2222 · 1m 9s · 15 turns · 22 in / 3,784 out tokens

With context: $0.2983 · 1m 19s · 16 turns · 1,173 in / 4,266 out tokens

95%

55%

Customer Call Analytics Ingestion

JSON-format transcription for data pipelines

Criteria

Without context

With context

Uses transcribe.sh

100%

No direct curl call

100%

JSON flag used

100%

Output path flag

100%

Output extension is .json

100%

API key from environment

100%

openclaw config mentioned

Audio file as argument

100%

call_transcripts/ directory created

100%

Without context: $0.1823 · 48s · 13 turns · 20 in / 2,989 out tokens

With context: $0.2802 · 1m 2s · 18 turns · 277 in / 3,766 out tokens

Repository: trpc-group/trpc-agent-go
Commit: 8763418

Evaluated: 16 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Podcast Transcription Automation Field Interview Transcription Pipeline Customer Call Analytics Ingestion

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.