agentleFS
beratmemisogluberatmemisoglu/claude-viral-animasyon-sablonu/skills/competitor-radar/SKILL.md

competitor-radar

Skill4 starsChanged 13 days agoMITView on GitHub

Daily competitor intelligence for the faceless animated YouTube channel "The Other Self". Scans the competitor channels with the YouTube Data API (what is being watched fast right now, outliers vs each channel's median, durations, title/thumbnail patterns, upload cadence), stores daily snapshots to measure 24h view velocity, then turns the evidence into ORIGINAL film ideas for the inner-voice-film skill (one pain word title + hook + fresh visual metaphor) — never copies a competitor's title, story, character or thumbnail. Use it every day before choosing the next film, and whenever the owner asks "rakipler ne yapıyor", "rakip analizi", "trend", "içerik fikri", "ne izleniyor", "hangi konu tutar".

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    write project/films/<slug>/BRIEF.md for today's pick (title, hook, ritual, voice lie ladder seed, metaphor, evidence) — this is the input …

    line 46
  2. Very specific:

    Learn: if our own videos have analytics (studio/scripts/youtube/yt_stats.py), compare them with the competitors' winners and add 1–3 …

    line 49

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Competitor Radar — The Other Self

Goal: every day, learn what the audience of faceless "inner struggle" animation is consuming fastest and convert it into one original film decision (+ a short ranked backlog) for inner-voice-film.

Competitors (edit competitors.json)

Day One Films (main), IfverseTales, Last Soul Origin, cheslimpico, Little Wars Films, Inner War Films, Madness Paint, Pointless Fiction, If Humans.

Daily procedure (do every step, in order)

  1. Scan (≈100 API quota units, free): ~/the-other-self/.venv-yt/bin/python ~/.claude/skills/competitor-radar/scripts/competitor_scan.py → ~/the-other-self/project/competitors/data/YYYY-MM-DD.json (raw snapshot) and ~/the-other-self/project/competitors/YYYY-MM-DD.md (metrics report: 24h velocity leaders, outliers, newest uploads, title patterns, durations, cadence, Shorts vs long). Read the report fully.
  2. Look at the winners' thumbnails (vision): scripts/competitor_scan.py downloads the top thumbnails to project/competitors/thumbs/YYYY-MM-DD/. Open the top 6 with the vision tool and note the craft only: emotion, colour, text (words + punctuation), composition, what is impossible/surprising. Never pass these images to Higgsfield or any generator (rule: competitor thumbnails are analysis-only).
  3. Diagnose with references/method.md §1: for each top performer write the underlying pain (one word), the format signal (length, title form, hook type), and why it travels (identity, universality, curiosity). Separate signal from noise: velocity and outlier score beat raw views; one viral video ≠ a trend (need ≥2 channels or ≥2 videos).
  4. Generate ideas with §2: at least 6 idea cards, each = TITLE (one ALL-CAPS pain word people say about themselves) · hook line · ritual · the twin's lie · fresh visual metaphor · evidence (which competitor data) · originality check.
  5. Originality gate (§3, mandatory): reject any idea whose title word, story spine, metaphor or set already exists in project/films/PUBLISHED.md, project/TOPICS.md [x] items, or a competitor title from the last 60 days. Different angle on the same pain is allowed only with a new metaphor and a new ritual.
  6. Score & pick (§4 rubric, 0–5 each: demand, velocity trend, originality, metaphor strength, universality, production fit). Pick one film for today; keep the rest in project/TOPICS.md "Next" (newest evidence first).
  7. Write outputs:
    • append a dated section to project/competitors/IDEAS.md (all cards + scores),
    • update project/TOPICS.md,
    • write project/films/<slug>/BRIEF.md for today's pick (title, hook, ritual, voice lie ladder seed, metaphor, evidence) — this is the input for inner-voice-film,
    • one-paragraph Turkish summary for the owner (what is hot, today's pick, why).
  8. Learn: if our own videos have analytics (studio/scripts/youtube/yt_stats.py), compare them with the competitors' winners and add 1–3 concrete lessons to project/LEARNINGS.md.

Hard rules

  • Copy the logic, never the surface: no competitor titles, lines, characters, plots, thumbnails, colours (Day One Films' red, their bean character).
  • Our universe only: Moon head + his blue shadow twin, black letterboxed stage, ~2:45, ElevenLabs twin voice.
  • Evidence or it didn't happen: every idea cites numbers from the snapshot. Never invent metrics; if the scan fails, say so and fall back to project/TOPICS.md.
  • Quota: one scan per day is enough (≈100 units of 10,000).

Files

FileUse
competitors.jsonhandles to scan (add/remove here)
scripts/competitor_scan.pyAPI scan → snapshot JSON + metrics report + top thumbnails
references/method.mddiagnosis → idea cards → originality gate → scoring rubric, with examples

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.