تشغيل أي مهارة في Manus بنقرة واحدة

scout-sources

النجوم٢

التفرعات٠

آخر تحديث١٠ يوليو ٢٠٢٦ في ٠٤:٠٥

Deterministic source-video discovery — the clip-channel "outlier" method, keyless. Searches seed niche queries via yt-dlp ytsearch, prefilters to long-form candidates, fetches full metadata, and ranks by outlier score (views/day velocity, views-per-subscriber ratio, comment engagement, replay-heatmap peakiness). Emits a ranked candidates.json to pick the next pipeline source from. No Claude call, no API key.

التثبيت

التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.

تشغيل في Manus

المصدر

jperrello

jperrello/C0BALT_CUT

فتح مستودع GitHub عرض مستودعات المنشئ

تنزيل

تشغيل في Manus

المهن ذات الصلةSOC

استنادا إلى تصنيف SOC المهني

محررو الأفلام والفيديوالفنون والتصميم والترفيه والرياضة والإعلام·SOC 27-4032

مستكشف الملفات

4 ملفات

SKILL.md

readonly

name	scout-sources
description	Deterministic source-video discovery — the clip-channel "outlier" method, keyless. Searches seed niche queries via yt-dlp ytsearch, prefilters to long-form candidates, fetches full metadata, and ranks by outlier score (views/day velocity, views-per-subscriber ratio, comment engagement, replay-heatmap peakiness). Emits a ranked candidates.json to pick the next pipeline source from. No Claude call, no API key.
allowed-tools	Bash
user-invocable	true

scout-sources

Answers "what video should we clip next?" deterministically instead of by vibes. Professional clip channels hunt outliers — videos performing far above their channel's baseline. The cheap keyless proxies, all available from yt-dlp:

velocity 2 * log10(1 + views/day) — reach × freshness
outlier min(10, views/subscribers) — 3x+ is real signal, 5–10x is strong
engagement min(5, comments per 1k views) — talked-about-ness
peaky min(3, 10 * stdev(replay heatmap)) — a spiky most-replayed graph means the video CONTAINS clippable moments; flat means evenly mediocre
curiosity min(1.5, question/superlative framing in the title) — sources whose own title is a curiosity hook ("how come…", "richest woman ever") yield the cold-open question moments the pipeline now leads with

score = velocity + outlier + engagement + peaky + curiosity (see score.py).

Invoke

.claude/skills/scout-sources/scout-sources.sh [out.json] [query ...]

out.json (optional): default work/_scout/candidates.json
query ... (optional): seed searches; default reads niches.txt (channel analytics — humor pods + productivity/AI — plus visually-rich, question-driven niches: space, wealth, true-crime, nature, history)

Env knobs: SCOUT_PER_QUERY (12 results/query), SCOUT_SHORTLIST (20 full fetches), SCOUT_MIN_VIEWS (100000), SCOUT_DUR_MIN/SCOUT_DUR_MAX (900/10800s — long-form only).

Output

candidates.json with ranked candidates (url, title, channel, score + components, seen: true when work/<sha1(url)[:10]> already exists) and a ranked table on stdout. Feed the winner straight to start.sh <url>.

How

Parallel flat ytsearch per query — view_count/duration/channel only (one cheap call per niche).
Merge, dedup, drop live/short/low-view, keep top-N by views.
4-way-parallel full yt-dlp -J per shortlisted id — adds upload_date, channel_follower_count, comment_count, heatmap.
score.py ranks and writes candidates.json.

NOT part of the per-video pipeline — run it before start.sh to decide what to feed the pipeline. Uses plain yt-dlp ytsearch (NEVER mcptube discover, which needs a forbidden LLM API key).

المزيد من هذا المستودع

نفس المستودع

title-transition

jperrello/C0BALT_CUT

Animate the title as a COLD OPEN in the TOP banner over the live opening footage — no blocking center card — held until TITLE_SWAP (default 2.0s) so brand-overlays can hand off to the source citation in the same slot. Frame 1 stays content (a face mid-sentence) to protect the swipe-rate / "Stayed to Watch" signal. ONE animation for every short — the glitch style (RGB-split banner + scanline tear, style-matched zap/crackle SFX) — a PIL frame sequence anchored at TITLE_ANCHOR_FRAC (default 0.135).

2026-07-172

generate-title

jperrello/C0BALT_CUT

Generate a per-clip third-person engagement-driven title card text. Claude reads the clip's transcript, the source ingest.json metadata, and (optionally) pick-segments' per-span judgment (topic, rationale, suggested title) so it reads the speaker's register — sincere vs ironic vs joking — before titling. Emits a hook-driven ALL-CAPS title (<=7 words).

2026-07-142

grade-clip

jperrello/C0BALT_CUT

Per-clip upload-readiness grade (0-99) read off a FINISHED .mp4 plus its persisted sidecar plans — the first skill that inspects a delivered pixel-level artifact, the on-disk proxy for YouTube's VVSA swipe gate. Deterministic retention-proxy floor (frame1_is_face, letterbox_bars, credit_lit_at_open, first_visual_change, first_payoff_offset, longest_static_gap, opening_caption_words, max_residual_silence, terminal_loop_score) with hard caps (letterbox / face_withheld / credit_at_open / blocking_card / dead_tail) that cap grade≤40, plus ONE batched Claude rubric call (hook↔payoff / open-loop / cold-context) skipped when GRADE_SKIP_CLAUDE=1. Runs at the END of the per-span chain (after save-local) AND standalone over the whole output/ backlog (--backlog → output/_triage.json). Emits clip_NN.grade.json per the locked SELECTION-SUITE-CONTRACT schema. Idempotent (.gcmeta), non-fatal (any error → DROSS/empty, exit 0).

2026-07-142

broll-pick

jperrello/C0BALT_CUT

Claude picks CONTEXTUAL/scene-level anchors from a clip transcript (literal objects, establishing shots, evocative concept footage matched to the story's tone — not just keyword nouns) and sources full-bleed B-roll cutaways via keyless yt-dlp YouTube search + mcptube frame-sampling + Claude vision verify. The verify step gets the spoken context and rejects literal-but-wrong matches (e.g. a cat laser toy for a tense "red dot" beat). Aims dense (8-12 windows, targeting cutaway footage on ~60%+ of the runtime; emits cutaway_coverage in the plan and WARNs below BROLL_MIN_COVER, default 0.6). Emits broll_plan.json with per-cutaway placement [t0,t1] and source metadata; clip files are namespaced per-clip. Windows snap to whole chunk-captions boundaries; vision calls capped by BROLL_VISION_CAP (default 16).

2026-07-102

pick-segments

jperrello/C0BALT_CUT

Pick N clip-worthy spans from a transcript. Claude reads the full transcript plus a per-second RMS energy summary and returns the most engaging time ranges for shorts. The atomic unit of a short is ONE IDEA, not one topic — each span carries `cuts` (1-3 non-contiguous source ranges assembled into one flowing story) plus `idea` (the one-sentence throughline all cuts share). A multi-cut span may pull cuts from DIFFERENT topics when they develop the same single idea; a single-cut span must stay within one topic and is just `cuts:[[t0,t1]]`. A 1-second SPOKEN hook is mandatory (no title card rescues a slow open — picks under PICK_HOOK_FLOOR are dropped). Replaces the old keyword/hook-word heuristic ranker.

2026-07-102

verify-bookends

jperrello/C0BALT_CUT

Post-edit verification of a short's opening and closing 1.5s. Claude sees a 3-frame strip from each end plus the trimmed transcript text around each end and returns {action:keep}, {action:trim,t0,t1} (inward-only), or {action:drop,reason}. Checks three things — (1) cleanliness (partial words, breath cutoffs, off-shot frames); (2) opening-hook strength (snap t0 inward to a stronger hook line if the first 3s is pure setup); (3) payoff landing (snap t1 to ~80ms past the payoff word when the tail trails into filler). Runs after tighten-pace, before fit-vertical. Inward-only — bookend-trim's outward pass already had its chance. Drops the span only on cleanliness failures requiring >2s of trim — never on hook-weakness alone.

2026-07-102

name	scout-sources
description	Deterministic source-video discovery — the clip-channel "outlier" method, keyless. Searches seed niche queries via yt-dlp ytsearch, prefilters to long-form candidates, fetches full metadata, and ranks by outlier score (views/day velocity, views-per-subscriber ratio, comment engagement, replay-heatmap peakiness). Emits a ranked candidates.json to pick the next pipeline source from. No Claude call, no API key.
allowed-tools	Bash
user-invocable	true

scout-sources

velocity 2 * log10(1 + views/day) — reach × freshness
outlier min(10, views/subscribers) — 3x+ is real signal, 5–10x is strong
engagement min(5, comments per 1k views) — talked-about-ness
peaky min(3, 10 * stdev(replay heatmap)) — a spiky most-replayed graph means the video CONTAINS clippable moments; flat means evenly mediocre
curiosity min(1.5, question/superlative framing in the title) — sources whose own title is a curiosity hook ("how come…", "richest woman ever") yield the cold-open question moments the pipeline now leads with

score = velocity + outlier + engagement + peaky + curiosity (see score.py).

Invoke

.claude/skills/scout-sources/scout-sources.sh [out.json] [query ...]

out.json (optional): default work/_scout/candidates.json
query ... (optional): seed searches; default reads niches.txt (channel analytics — humor pods + productivity/AI — plus visually-rich, question-driven niches: space, wealth, true-crime, nature, history)

Env knobs: SCOUT_PER_QUERY (12 results/query), SCOUT_SHORTLIST (20 full fetches), SCOUT_MIN_VIEWS (100000), SCOUT_DUR_MIN/SCOUT_DUR_MAX (900/10800s — long-form only).

Output

How

Parallel flat ytsearch per query — view_count/duration/channel only (one cheap call per niche).
Merge, dedup, drop live/short/low-view, keep top-N by views.
4-way-parallel full yt-dlp -J per shortlisted id — adds upload_date, channel_follower_count, comment_count, heatmap.
score.py ranks and writes candidates.json.

NOT part of the per-video pipeline — run it before start.sh to decide what to feed the pipeline. Uses plain yt-dlp ytsearch (NEVER mcptube discover, which needs a forbidden LLM API key).