obra/ghost-pepper
Hold-to-talk speech-to-text for macOS. 100% local, powered by WhisperKit and local LLM cleanup. Hold Control to record, release to transcribe and paste.
https://skillcdn.ai/gh/obra/ghost-pepperConnect this address to your AI to use these skills. How to connect
- Unverified
- Default branch
main - Commit
a9eed76 - License
None declared
Explore this repository
Searches the original text of skills and documents; display translations are not searched.
Documents
- Onboarding Flow Implementation PlanGoal: Build a first-run onboarding wizard that walks users through permissions, model download, and a live try-it step before they use GhostPepper.docs/superpowers/plans/2026-03-20-onboarding.md
- Transcription Stack Implementation PlanGoal: Build configurable recording chords, cleanup backend/model selection, OCR-informed cleanup, and conservative correction learning as a stacked series of feature branches.docs/superpowers/plans/2026-03-23-transcription-stack.md
- Cleanup Model Probe Implementation PlanGoal: Build a local CLI probe that loads Ghost Pepper's real cleanup models and shows prompt, raw output, sanitized output, and final cleaned output for exact inputs.docs/superpowers/plans/2026-03-25-cleanup-model-probe.md
- FluidAudio STT Backend Implementation PlanGoal: Add a unified speech model picker that routes internally to WhisperKit or FluidAudio so GhostPepper can offer Parakeet v3 without exposing backend details.docs/superpowers/plans/2026-03-25-fluid-audio-stt.md
- Settings Sidebar And Sound Toggle Implementation PlanGoal: Add a single Play sounds preference and rebuild Ghost Pepper's settings window into a large, spacious, macOS-style sidebar settings app without removing any existing settings.docs/superpowers/plans/2026-03-25-settings-sidebar-and-sound-toggle.md
- Transcription Lab Implementation PlanGoal: Add a Settings-based Transcription Lab that archives the last 50 non-empty recordings and lets users rerun them with different speech models, cleanup models, and cleanup prompt text using the o…docs/superpowers/plans/2026-03-25-transcription-lab.md
- FluidAudio Diarization Implementation PlanGoal: Add a Ignore other speakers recording preference that uses FluidAudio diarization to keep the first substantial speaker in stop-and-paste dictation, while remaining visible but disabled for Whi…docs/superpowers/plans/2026-03-26-fluidaudio-diarization.md
- Paste Fallback Implementation PlanGoal: Fall back to leaving transcribed text on the clipboard and telling the user to press Cmd-V when Ghost Pepper cannot confidently paste into the current target.docs/superpowers/plans/2026-03-28-paste-fallback.md
- Context Bundler Opt-In Implementation PlanGoal: Make Context Bundler opt-in so new users do not see or trigger it until they enable it, while preserving the existing enabled behavior for users who already stored a Zo token.docs/superpowers/plans/2026-04-09-pepper-chat-opt-in.md
- Speaker Identities Implementation PlanGoal: Add durable recognized voices for transcription-lab speaker tagging, with local/global naming controls and speaker labels directly on the lab timeline.docs/superpowers/plans/2026-04-20-speaker-identities.md
- Agentic Meeting Q&A Implementation PlanGoal: Replace Ghost Pepper's keyword-scoring + 30 KB-cram cross-meeting Q&A core with a bounded agentic tool-use loop (grep/read_file/list_dir, cap 15 iterations) over the local meeting archive, behi…docs/superpowers/plans/2026-04-28-agentic-meeting-qa.md