Manages OpenWhispr voice transcripts and watermarking for idempotent processing of notes.
npx skills add https://github.com/philipyaz/cos --skill voice-operations
Please help me install this Agent Skill. Skill: voice-operations Source: https://github.com/philipyaz/cos/tree/main/.claude/skills/voice-operations Command: npx skills add https://github.com/philipyaz/cos --skill voice-operations
This Skill streamlines the management of OpenWhispr voice transcripts, allowing for efficient processing and idempotent handling of voice notes.
Use the voice-operations skill to list all unprocessed transcripts from OpenWhispr.
High-intent search queries and answers about installing and using this skill.
Transcript watermarking enables idempotent processing of voice notes by tracking the last processed record. This ensures each voice note is handled exactly once, preventing duplicate entries in your workflow.
Managing unprocessed OpenWhispr voice notes involves reading available transcripts and applying a watermark for idempotent processing. This approach filters out already handled voice notes, ensuring only new audio files are queued for processing.
Voice transcription watermarking works by placing a processing marker on each transcript. Once a voice note is handled, the watermark records this state, ensuring the system skips it during future operations to maintain idempotency.
Yes, you can list all unprocessed voice transcripts from OpenWhispr in a single operation. The system reads available transcripts and filters them against the current watermark to return only pending voice notes.
No external dependencies are required to manage OpenWhispr voice transcripts. The skill operates independently within the OpenWhispr ecosystem to read transcripts and handle idempotent processing watermarks.
Idempotent processing for voice notes is needed when transcripts are handled in batches or recurring workflows. It ensures that repeated runs do not duplicate work, keeping already processed audio files from being reprocessed.
philipyazAutomate OpenWhispr voice notes MCP setup for Claude Code and Cowork clients.
pnils08Generate structured two-host podcast transcripts as XML from edition stories.
ma08Convert audio and video files into markdown transcripts with speaker labels.
xingBaGanAnalyze video content and transcribe audio locally with ffmpeg and Whisper.
