recoup-short-video
End-to-end async run that produces a 9:16 social-ready short-form music video (TikTok/Reel/Short) featuring an artist and their song. Use whenever the user types `/recoup-short-video`, says "make a video for [artist]", "create a TikTok for [artist]", "produce a Reel for [artist]", "kick off content for [artist]", or any front-door request to generate a finished short-form clip for an existing artist. Resolves the artist's `account_id`, fires `POST /api/content/create`, polls `/api/tasks/runs` until terminal, and lands the user on the final video URL + caption. The default front door for the recoup-content plugin.
What this skill does
# Recoup Short Video
The anchor skill for the recoup-content plugin. Use this on first install to confirm the plugin works end-to-end, and as the default front door for "make a video for [artist]" requests.
This skill is **user-invoked** — the user types `/recoup-short-video <artist-name>` and the harness invokes it with `$ARGUMENTS` populated. It is self-contained: the default async path below runs without any other skill. For granular control (swapping a single stage), it points to the `short-video` skill in the Recoupable skills library.
## Arguments
The user invoked this with: `$ARGUMENTS`
Expected shape: `<artist-name> [--template <template-name>]`
- **`<artist-name>`** — the artist to generate content for. Required. If the user didn't supply one, AskUserQuestion to collect it.
- **`--template <name>`** — optional. Defaults to `artist-caption-bedroom`. Override based on user cue ("with the album cover" → `album-record-store`, etc.).
## Required environment
- `RECOUP_API_KEY` set in the shell environment, OR a Privy access token if running inside chat (pass it as `-H "Authorization: Bearer $RECOUP_ACCESS_TOKEN"` instead of `x-api-key`).
- An artist record already exists for the requested artist on the user's first org. If not, run the `recoup-create-artist` skill (Recoupable skills library) first.
## What it does
1. Resolves the artist's `account_id` (from artist name + the user's first org).
2. Picks a template (defaults to `artist-caption-bedroom` if not supplied).
3. Fires `POST /api/content/create` to start the async pipeline server-side.
4. Polls `/api/tasks/runs?runId={runId}` every ~10 seconds until status is `COMPLETED`, `FAILED`, `CANCELED`, or `CRASHED`.
5. Reads the run output: `videoSourceUrl`, `imageUrl`, `captionText`, `template`, `lipsync`, and the audio metadata.
6. Lands the user on the final video URL with the caption text printed below it.
The async pipeline is the agent-safe path — synchronous calls to `/api/content/video` routinely take 60–180 seconds and time out inside most agent shells.
## Template looks (studio / stage / bedroom / …)
The artist's *look* is the **template** — same video job, different aesthetic preset. Don't
make a separate request type per vibe; pick a template. Known looks:
| Look | Template |
| --- | --- |
| Bedroom / intimate | `artist-caption-bedroom` |
| On stage / performance | `artist-caption-stage` |
| Outside / street | `artist-caption-outside` |
| Album in a record store | `album-record-store` |
Templates evolve — list the live set rather than trusting this table:
```bash
curl -sS "${AUTH[@]}" "https://api.recoupable.com/api/content/templates" \
| jq -r '.templates[] | "\(.id) — \(.description)"'
```
Pick from the user's cue ("on stage" → `artist-caption-stage`, "with the album cover" →
`album-record-store`); default `artist-caption-bedroom`. Full details:
`references/content-api.md`.
## Context vs generic mode
- **Context mode** (artist named, workspace exists): pull the reference image from
`context/images/face-guide.png` and aesthetic from `context/artist.md`; resolve the song
from the workspace. See `references/workspace-context.md`.
- **Generic mode** (no artist context wanted): the user supplies the reference image,
prompt, and audio directly — run the same pipeline without artist context. Don't force a
workspace.
## Steps the agent must execute
### 1. Resolve the artist's `account_id`
```bash
ORG_ID=$(curl -sS "https://api.recoupable.com/api/organizations" \
-H "x-api-key: $RECOUP_API_KEY" | jq -r '.organizations[0].id')
ARTIST_ACCOUNT_ID=$(curl -sS "https://api.recoupable.com/api/artists?org_id=$ORG_ID" \
-H "x-api-key: $RECOUP_API_KEY" \
| jq -r --arg name "$ARTIST_NAME" '.artists[] | select(.name == $name) | .account_id')
```
Use **`account_id`**, not `id` — `id` is the artist row's primary key, `account_id` is the underlying account that owns it. Passing the wrong one returns a 404 from `/api/content/create`. If no match, the artist record doesn't exist yet — run the `recoup-create-artist` skill first.
### 2. Trigger the async pipeline
```bash
TEMPLATE="${TEMPLATE:-artist-caption-bedroom}"
RUN_IDS=$(curl -sS -X POST "https://api.recoupable.com/api/content/create" \
-H "x-api-key: $RECOUP_API_KEY" -H "Content-Type: application/json" \
-d "$(jq -n --arg artist "$ARTIST_ACCOUNT_ID" --arg template "$TEMPLATE" \
'{artist_account_id: $artist, template: $template}')" \
| jq -r '.runIds[]')
RUN_ID=$(echo "$RUN_IDS" | head -1)
```
### 3. Poll until terminal
```bash
until STATUS=$(curl -sS "https://api.recoupable.com/api/tasks/runs?runId=$RUN_ID" \
-H "x-api-key: $RECOUP_API_KEY" \
| jq -r '.runs[0].status') && \
[[ "$STATUS" =~ ^(COMPLETED|FAILED|CANCELED|CRASHED)$ ]]; do
sleep 10
done
```
### 4. Read the output and land the user
```bash
curl -sS "https://api.recoupable.com/api/tasks/runs?runId=$RUN_ID" \
-H "x-api-key: $RECOUP_API_KEY" \
| jq '.runs[0].output'
# -> { videoSourceUrl, imageUrl, captionText, template, lipsync, audio: {...} }
```
See [Tasks Runs](https://developers.recoupable.com/api-reference/tasks/runs) for the full status enum and the `CreateContentRunOutput` schema.
## What "complete" looks like
The command finishes when:
- The poll loop has seen a terminal status (`COMPLETED` is the happy path).
- `output.videoSourceUrl` is a fetchable URL.
- `output.captionText` is a non-empty string.
Print both to the user. If status is `FAILED`, `CANCELED`, or `CRASHED`, surface the error from `runs[0].error` and stop — don't claim success.
**Before claiming success, run the analyze gate.** The agent can't see pixels — verify the
render isn't glitchy, empty, or off-brand via `references/analyze-gate.md` rather than
trusting a `COMPLETED` status alone.
For the underlying audio (lipsync templates), see `references/song-sourcing.md` — never
substitute a placeholder track.
## When to reach for the other skills
This async front door is the default. Use the underlying skills directly when:
- The user wants to **swap a single stage** (different caption length, different motion prompt, different reference image), inspect intermediate outputs, or supply their own audio. Use the **`short-video`** skill (Recoupable skills library) — its bundled manual covers the five-step recipe, and it explains resolving the underlying `song.mp3`.
- The user wants to **generate one capability in isolation** (just an image, just a caption). Use the **`content-creation`** skill (Recoupable skills library).
- The artist **doesn't exist yet**. Use the **`recoup-create-artist`** skill (Recoupable skills library) first.
## References
Shared building blocks (ship alongside this skill):
- `references/content-api.md` — endpoints, the 6 video modes, templates, async create→poll.
- `references/account-resolver.md` — auth modes + `account_id` vs row `id`.
- `references/workspace-context.md` — context vs generic mode; read context, write back.
- `references/song-sourcing.md` — sourcing the song audio (no placeholders).
- `references/analyze-gate.md` — verify the render before claiming done.
Related in Ads & Marketing
ads
IncludedMulti-platform paid advertising audit and optimization skill. Analyzes Google, Meta, YouTube, LinkedIn, TikTok, Microsoft, and Apple Ads. 250+ checks with scoring, parallel agents, industry templates, and AI creative generation.
banana
IncludedAI image generation Creative Director powered by Google Gemini Nano Banana models. Use this skill for ANY request involving image creation, editing, visual asset production, or creative direction. Triggers on: generate an image, create a photo, edit this picture, design a logo, make a banner, visual for my anything, and all /banana commands. Handles text-to-image, image editing, multi-turn creative sessions, batch workflows, and brand presets.
rpg-migration-analyzer
IncludedAnalyzes legacy RPG (Report Program Generator) programs from AS/400 and IBM i systems for migration to modern Java applications. Extracts business logic from RPG III/IV/ILE source code, identifies data structures (D-specs), file operations (F-specs), program dependencies (CALLB/CALLP), and converts RPG constructs to Java equivalents. Generates migration reports, complexity estimates, and Java implementation strategies with POJO classes, JPA entities, and service methods. Use when modernizing AS/400 or IBM i legacy systems, analyzing RPG source files (.rpg, .rpgle, .RPGLE), converting RPG to Java, mapping data specifications to Java classes, planning legacy system migration, or when user mentions RPG analysis, Report Program Generator, RPG III/IV/ILE, AS/400 modernization, IBM i migration, packed decimal conversion, or mainframe application rewrite.
brand-library-architect
IncludedBuild a complete brand library for a product — visual asset render pipeline, brand documentation set (BRAND, COPY, MANIFESTO, BIOS, FAQ, GLOSSARY, TONE, PRICING), open-source convention files (README, CONTRIBUTING, SECURITY, CODE_OF_CONDUCT), and a self-contained press kit. This skill should be used when the user asks to "build a brand library / brand kit / press kit / brand assets" for a product, "set up a brand library workflow," "create a positioning manifesto plus visual identity," or any combination of brand documentation + visual asset pipeline. Apply phase-by-phase or run end-to-end. Templates are product-agnostic and use {{TOKEN}} placeholders the skill prompts the user to fill.
writing-tech-post
IncludedAuthors engineering blog posts end-to-end: launch deep-dives, incident postmortems, architecture migrations, performance case studies, tutorials, AI/agent system writeups, security disclosures, and research-to-product translations. Picks the correct archetype, plans the abstraction ladder, enforces an evidence cadence (diagrams, benchmarks, profiles, traces, code, ablations), tunes voice against publisher house styles (Datadog, Vercel, GitHub, AWS, Meta, Cloudflare, Jane Street), and runs a pre-publish gate for narrative momentum and disclosure ethics. Use when drafting a new engineering post, restructuring a draft that feels flat, deciding which evidence form belongs where, validating that depth and product context are balanced, or preparing a postmortem, migration, or performance narrative for external publication. Do not use for API reference documentation, README authoring, marketing copy, release notes, generic SEO content, ghost-written executive thought leadership, or non-engineering long-form essays.
blog-google
IncludedGoogle API integration for blog performance: PageSpeed Insights, CrUX Core Web Vitals with 25-week history, Search Console performance, URL Inspection, Indexing API, GA4 organic traffic, NLP entity analysis for E-E-A-T, YouTube video search for embedding, and Google Ads Keyword Planner. Progressive feature availability based on credential tier (API key, OAuth/service account, GA4, Ads). Shares config with claude-seo at ~/.config/claude-seo/google-api.json. Use when user says "google data", "page speed", "core web vitals", "search console", "indexation", "GA4", "keyword research", "nlp entities", "blog performance", "youtube search", "google api setup".