Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/higgsfield-ai/cursor-pluginWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/higgsfield-ai/cursor-plugin/higgs)<a href="https://agentmods.dev/commands/higgsfield-ai/cursor-plugin/higgs"><img src="https://agentmods.dev/badge/commands/higgsfield-ai/cursor-plugin/higgs/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/higgsfield-ai/cursor-plugin/higgs"><img src="https://agentmods.dev/badge/commands/higgsfield-ai/cursor-plugin/higgs.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00024 | $0.00494 |
| Opus 5.5 | $0.00010 | $0.00198 |
| Sonnet 5.5 | $0.00005 | $0.00099 |
| Haiku 4.5 | $0.00002 | $0.00049 |
Grade A, and why
higgs scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/higgs
Natural-language entrypoint for Higgsfield. Interpret the request, preserve the user's constraints, and use the current Higgsfield MCP tool descriptions as the source of truth.
Usage
/higgs <natural language request>
Examples
/higgs Make a cinematic poster of a samurai in neon Tokyo, 16:9
/higgs Animate this product shot — slow push-in, 5s, 9:16
/higgs Make a 60-second faceless explainer about black holes
/higgs Score this video for engagement and tell me what to fix
/higgs Show my recent generations
/higgs How many credits do I have left?
Routing
- One image, video, or audio result → use
generate_image,generate_video, orgenerate_audiodirectly. Usemodels_explorewhen the model or its parameters are unclear. - Several independent generations → use the matching batch generation tool and follow its
jobs_waitand display protocol. - Specialized production → call
get_workflow_instructionswith no workflow first to discover the current catalog, then load and follow the matching canonical workflow. This includes faceless videos, UGC formats, narration, subtitles, thumbnails, brand kits, and character sheets. - Edit an existing asset → prefer the dedicated operation when available, including upscale, outpaint, reframe, background removal, motion control, voice change, or dubbing.
- Virality or audience-response analysis → use the MCP virality analysis tool selected by its current description.
- Library, characters, workspaces, billing, and account requests → use the matching MCP browse or account tool.
Media inputs
- For an HTTP(S) media URL, import it before passing the returned media ID to another tool.
- For a local file, use the upload flow exposed by the current MCP client and confirm the upload when required.
- Preserve every supplied reference and its intended role.
Do not turn a generic generation request into a complex workflow. Do not invent tool names, model IDs, parameters, workflow steps, or defaults when MCP can provide them. Follow typed recovery guidance instead of retrying an unchanged failed call.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 43 lines · 24 tokens per session scan A e313fe271037
higgs is a command published in the GitHub repository higgsfield-ai/cursor-plugin (37 stars, last pushed 2mo ago), licensed MIT. It adds 24 tokens to every session and 494 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
higgs compared with commands from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| higgshiggsfield-ai/cursor-plugin | Command | 37 | 2mo ago | MIT | A | 24 | Claude Code |
| checklistgithub/spec-kit | Command | 148k | yesterday | MIT | A | 13 | Claude Code |
| clarifygithub/spec-kit | Command | 148k | yesterday | MIT | A | 27 | Claude Code |
| specifygithub/spec-kit | Command | 148k | yesterday | MIT | A | 13 | Claude Code |
| analyzegithub/spec-kit | Command | 148k | yesterday | MIT | A | 26 | Claude Code |
| convergegithub/spec-kit | Command | 148k | yesterday | MIT | A | 35 | Claude Code |
| implementgithub/spec-kit | Command | 148k | yesterday | MIT | A | 14 | Claude Code |