207studio
OrganizationExperimental opt-in Jev skills and config for Claude Code — the companion to jev-codex-tools.
Categories
Indexed Skills (6)
jev-code-search
Use Jev for broad code relevance searches likely to return many passages. Use bounded native rg for known files, exact symbols, and small tasks. Applies relevance filtering before loading code into context.
jev-voice
Build or tune a voice-command UI where Jev turns each utterance into typed decisions — what to do, on what, and whether the speaker has finished — while code executes. Use for speech-driven app control, choosing on-screen buttons by voice, end-of-speech detection, or wiring Apple's SpeechTranscriber to Jev. Agent-neutral (Claude Code and Codex).
jev-code-sweep
Sweep a whole repository or large change for likely runtime defects by having Jev triage every ~100-line chunk before any agent reads code, then send only the top clusters to reviewers. Use for "review the whole codebase", pre-release defect sweeps, or when a review would otherwise load many files into context.
jev-action-control
Run explicitly authorized repeated browser, macOS or iOS actions with bounded Jev choices.
jev-mode
Batch 5+ fixed-label judgments without loading record text into agent context.
jev-verify
Decide whether an optional repeated verification is worth running.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.