lennoxsaint/Threadify-workflows / plugins/claude/skills/threadify-director-mode
threadify-director-mode
VerifiedStop editing. Start directing. Turn one raw talking-head video into a finished post in four levels - Speed (cut dead air and retakes), Altitude (move the strongest proof line to second 1), Depth (an outside judge scores the first 3 seconds, the agent fixes the top fault) and Range (sound effects, word-highlight captions and a title card) - with a one-line scoreboard after each, then Threadify writes the post in your voice and schedules it with the video only after one exact yes. Free and local; your original file is never touched. Use when someone has a raw clip, wants it edited for them, says "director mode", or asks to turn a talking-head take into a post.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 15.5 KB |
| references/generation-models.json | config | 751 B |
| references/post.md | markdown | 3.6 KB |
| references/threadify-001.md | markdown | 3.7 KB |
| references/workflow-manifest.json | config | 5.8 KB |
| references/workflow-readme.md | markdown | 1.9 KB |
| scripts/casing-guard.mjs | script | 7.1 KB |
| scripts/director-mode.mjs | script | 47.0 KB |
| scripts/director-net.mjs | script | 3.0 KB |
references/workflow-manifest.json
View on GitHub{
"workflow_id": "director-mode",
"version": "0.1.0",
"title": "Director Mode",
"summary": "Stop editing, start directing: turn one raw talking-head video into a finished post in four levels - cut dead air and retakes, move your strongest proof line to second 1, let an outside judge score the first 3 seconds and fix the top fault, then add free sound effects, word-highlight captions and a title card - with a one-line scoreboard after each; Threadify writes the post and schedules it with the video only after one exact yes. Free, local, and your original is never touched.",
"kind": "skill",
"entrypoint": {
"skill_id": "threadify-director-mode",
"source": "plugins/threadify/skills/threadify-director-mode/SKILL.md"
},
"bundle": {
"kind": "advanced",
"standalone": false
},
"dependencies": {
"workflows": [],
"files": [
"workflows/director-mode/README.md"
]
},
"triggers": [
"director mode",
"stop editing start directing",
"edit my raw video for me",
"turn my talking head video into a post",
"cut the dead air and retakes from my video",
"make my video something people finish"
],
"io": {
"inputs": [
{
"name": "one raw talking-head video file on this computer (60-90 seconds typical), ffmpeg, whisper.cpp with a speech model, optional Threadify connection for the post",
"schema_ref": "inline:director-mode-input"
}
],
"outputs": [
{
"name": "director/ folder beside the original with v1 (speed), v2 (altitude), v3 (depth) and final.mp4 (range, 1920x1080, at most 95 MB), a scoreboard line per level, two outside-judge scores, credits.json for sounds, verify result, and optionally one scheduled post with the video and its readback",
"schema_ref": "schemas/workflow-receipt.v1.json"
}
]
},
"supported_adapters": [
"generic-mcp",
"codex",
"claude",
"hermes",
"gemini",
"openclaw",
"cursor"
],
"required_mcp_tools": [
"get_connection_defaults",
"generate_content",
"create_media_upload",
"list_scheduled_posts",
"best_time_to_post",
"validate_post",
"schedule_post",
"get_schedule_status"
],
"optional_mcp_tools": [
"upload_media",
"get_brain_overview",
"query_brain",
"save_draft",
"get_draft"
],
"free_capabilities": [
"local word-timed transcription with whisper.cpp, one speech stretch at a time",
"dead-air cuts (over 0.3 s down to about 0.1 s) and retake removal that keeps the last clean take, with room tone kept at every join",
"reorder render: proof line to second 1 and cut lines, from a plan the agent writes",
"outside-judge packet (contact sheet, first-3-seconds frames and words, rubric) for a fresh subagent or external CLI",
"Openverse sound effects with a CC0/public-domain licence filter, rate-limit pacing, a local cache and credits.json",
"loudness-matched mix about 15 LU under the voice with cues kept off word onsets",
"burned word-highlight captions (lowercase, at most 3 words) and a title card (drawtext or a HyperFrames overlay)",
"final 1920x1080 30 fps MP4 under 95 MB, verified by full decode, format checks, re-transcription and cut-edge sound checks",
"local casing guard that proves lower(original) == lower(final) for the post"
],
"paid_capability_stubs": [
"optional Threadify post generation counts toward the plan's AI generations",
"optional connected video upload, scheduling and readback"
],
"approval_gate": {
"required": true,
"type": "explicit-final-approval",
"must_show": [
"account and timezone",
"the exact final post text and any sound credit line",
"the final video and its length",
"the local date and time",
"action: schedule this post with the video; nothing publishes now"
]
},
"fallback": {
"mode": "local-render-owner-posts",
"instructions": [
"Never write to the original video; every output goes to the director/ folder beside it.",
"Never upload the raw video, audio or transcript; only director/final.mp4 is uploaded, and only for the post.",
"When whisper-cli, a speech model or ffmpeg is missing, report the exact fix from the check command and run it with the owner's knowledge.",
"When ffmpeg lacks libass, finish without burned captions and say so.",
"When Openverse is down or over its limit, use cached sounds or none and say so; never use NC, ND or SA sounds, and CC BY only on opt-in with its credit line.",
"When no fresh subagent or external CLI is available, say the outside judge is unavailable and skip to Level 4; never judge your own edit as the outside judge.",
"When the judge's second score is not higher, say so plainly and keep the better version.",
"When verify finds a word missing at a cut that ran through sound, fix the cut before showing the final.",
"Never write the post yourself when Threadify generation is unavailable; report the blocker and the manual next step.",
"Never call publish_now; schedule only after the owner's exact yes to the post card, and read the schedule back.",
"When an upload or schedule outcome is unclear, read back before retrying and retry only with the same idempotency key."
]
},
"receipt": {
"schema": "workflow-receipt.v1",
"must_prove": [
"dependency check result and speech model used",
"original file untouched (outputs only in director/)",
"scoreboard line for each level",
"retakes cut and any vetoed",
"altitude order and reason",
"both outside-judge scores and the fault fixed",
"sound credits, licences and the measured LU under speech",
"verify result: plays, 1920x1080, 30 fps, size, words and cut edges",
"optional post: model, casing guard, upload, approval reply, scheduled slot and readback",
"fallback state"
]
}
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 469k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |