black-forest-labs-flux
Plan, prompt, execute, troubleshoot, and quality-control Black Forest Labs FLUX image generation and editing across the BFL direct API and licensed local/open-weight deployments. Use for FLUX.2 model selection, text-to-image, single- or multi-reference editing, typography, exact-color work, mask-based erase/inpainting, outpainting, API integration, reproducibility, deployment licensing, and production rights or safety decisions; also use when migrating or maintaining legacy FLUX.1/FLUX1.1 workflows.
Profile →npx skills add calesthio/generative-media-skills --skill black-forest-labs-flux --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Plan, prompt, execute, troubleshoot, and quality-control Black Forest Labs FLUX image generation and editing across the BFL direct API and licensed local/open-weight deployments. Use for FLUX.2 model selection, text-to-image, single- or multi-reference editing, typography, exact-color work, mask-based erase/inpainting, outpainting, API integration, reproducibility, deployment licensing, and production rights or safety decisions; also use when migrating or maintaining legacy FLUX.1/FLUX1.1 workflows.
How it works
- Identify the surface before selecting parameters or stating rights across: BFL direct API (api.bfl.ai with x-key), third-party hosted gateway, and local/open-weight deployment (weights/licensing). Verify provider docs and contracts for each surface. [Documented: D1, D2, D9] [Inference] [D10, D11]
- Choose the FLUX.2 path appropriate for your need (pro, max, flex, klein, dev) with described limits and licenses. [Documented: D1, D2, D10]
- For specialized operations (erase, outpainting, Fill paths, Kontext legacy paths), use the dedicated FLUX tools (Erase, Outpainting, Fill) and maintain awareness of licensing and current API contracts. [Documented: D7, D7, D7, D13, D9, D10]
- Before prompting, record contract details: final use, audience, endpoint, license, output specs, and rights. Stop if rights or consent are unresolved. [Documented: D1, D5, D12]
- Prompting: write natural-language prompts by front-loading priority content; support complex briefs with JSON-shaped strings when needed. Use quoted exact copy for critical text and specify placement/attributes in prompts. [Documented: D3, D5, D1, D2, D4]
- Direct API flow: POST JSON with prompt, optional input_image, dimensions, seed, moderation, format; capture id, polling_url, costs; poll until Ready; download asset with 10-minute signed URL; avoid logging keys or secrets. [Documented: D2, D5, D9]
- Use contracts for single/multiple references with clearly defined preservation constraints per image role; manage reference counts per current caps (8 for certain paths, 4 for klein; 6 recommended for dev). [Documented: D1, D6]
- Control typography, color, layout, and mask-driven operations with explicit operations: FLUX Erase, FLUX.1 Fill [pro], FLUX Outpainting, FLUX.2 editing with preserve lists. Include edge cases and halos handling per guidance. [Documented: D7, D5, D7, D5]
When to use it
Use when planning FLUX.2 workflows across surfaces (direct API, gateway, local weights) and when migrating/maintaining FLUX.1/FLUX1.1 workflows. Use specialized paths for erase, outpainting, and fill as appropriate to the task. [Documented: D1, D2, D7, D9, D10, D13]
What it can touch
- BFL direct API (api.bfl.ai) via x-key, including prompt, input_image, dimensions, seed, moderation, and format fields. [Documented: D2]
- Local/open-weight deployments and their licenses. [Documented: D10, D11]
- Reference images and prompt content as per contract rules. [Documented: D6]
Caveats
- Weights/licensing and endpoint behavior are volatile; recheck API references and licenses before consequential runs. [Documented: D1, D10, D12]
- Do not promise identical pixels across changing endpoints or weights. [Documented + Inference: D1, D2]
- Follow documented limits on references and parameters per path (e.g., up to 8 references for some paths, 4 for klein). [Documented: D1, D6]
# Produce images with Black Forest Labs FLUX Use this evidence notation throughout your work: - **[Documented]** means BFL documentation, a BFL model card/repository, BFL legal terms, or a BFL-authored paper states the fact. - **[Inference]** means a conclusion follows from documented facts but BFL does not state it directly. - **[Heuristic]** means production advice to validate against actual out
What does the black-forest-labs-flux skill do?
Plan, prompt, execute, troubleshoot, and quality-control Black Forest Labs FLUX image generation and editing across the BFL direct API and licensed local/open-weight deployments. Use for FLUX.2 model selection, text-to-image, single- or multi-reference editing, typography, exact-color work, mask-based erase/inpainting, outpainting, API integration, reproducibility, deployment licensing, and production rights or safety decisions; also use when migrating or maintaining legacy FLUX.1/FLUX1.1 workflows.
How do I install it?
Run `npx skills add calesthio/generative-media-skills --skill black-forest-labs-flux --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From calesthio/generative-media-skills, a repository with 112 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.