Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/OSideMedia/higgsfield-ai-prompt-skillnpx agentmods add skills/osidemedia/higgsfield-ai-prompt-skill/higgsfield-actingWrote 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/skills/osidemedia/higgsfield-ai-prompt-skill/higgsfield-acting)<a href="https://agentmods.dev/skills/osidemedia/higgsfield-ai-prompt-skill/higgsfield-acting"><img src="https://agentmods.dev/badge/skills/osidemedia/higgsfield-ai-prompt-skill/higgsfield-acting/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/skills/osidemedia/higgsfield-ai-prompt-skill/higgsfield-acting"><img src="https://agentmods.dev/badge/skills/osidemedia/higgsfield-ai-prompt-skill/higgsfield-acting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 89 Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.Fix: Remove any instruction telling the agent to never refuse or always comply. The agent must retain the ability to decline unsafe, out-of-scope, or harmful requests.
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 | $0.00165 | $0.08775 |
| Opus 5 | $0.00082 | $0.04387 |
| Sonnet 5 | $0.00033 | $0.01755 |
| Haiku 4.5 | $0.00016 | $0.00877 |
Grade A, and why
higgsfield-acting 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 13d 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 — 597 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Higgsfield Acting Director
[OFFICIAL — Higgsfield "Hell Grind" open-source brief] — the performance system behind
Higgsfield's 95-minute AI feature film, adapted to this repo's prompt surfaces.
The core axiom of the entire system: acting is BEHAVIOR under pressure, not a display of emotion. A character wants something, something is in the way, and they act to get it. Emotion is a byproduct of that struggle — never the thing you write directly.
This is a layer on top of ../higgsfield-seedance/SKILL.md: the paragraph this skill
produces goes into the PERFORMANCE / CHARACTER ACTING block of an otherwise normal Seedance
prompt. It does not change camera, light, wardrobe, or grade.
QUICK FACTS
Generated-checked block (scripts/build_index.py verifies anchors). Routing aids — read the linked sections for the rules.
- Write the objective (a verb aimed at a partner), never the state — "make him confess", never "be angry" →
- 2–4 beat changes per scene, each one visible in behavior: a pause, a posture change, a tempo change, a shift of gaze. Unchanged behavior for the whole shot = played flat →
- Listening beats speaking. The reaction starts before the partner's line ends; a hard question earns a micro-pause before the answer →
- The body carries the biography: center of gravity, tempo, openness, breath — set all four before any psychology →
- Business — a physical task the hands are doing — kills theatricality; the strongest accent in a scene is the moment the character stops it →
- Distance is drama; a change of distance is a beat change. Status is what you DO, and status breaks are the most interesting thing in a performance →
- The master profile is 150–220 words, one paragraph, fixed block order, written once per character and then rewritten per scene — never pasted →
- Every tic carries a trigger; every mask carries a crack — at least one "However, when X…" clause per profile →
- Eye life is mandatory and never optional — saccades, blink quality, live catchlights, eyes-lead-thought. Dead eyes are the number-one tell of AI acting →
- Scene adaptation transforms, never deletes: a behavior that can't physically happen is displaced into another outlet, not removed →
- The voice prompt is locked — one per character, pasted verbatim into the audio field, never adapted per scene →
- States, not transitions. Models fail process and nail state: "mid-throw, arm extended", not "reaches in, pulls out, winds up" →
- Ensemble reactions travel in a wave, never in sync; the strong are still and quiet, the weak fidget and shout →
- 15 named bad-acting symptoms with prompt-level fixes, and a 0–5 self-check scale — aim every hero shot at 4+ →
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.
- 13d ago First seen · 597 lines · 165 tokens per session scan A e76afce5b10f
higgsfield-acting is a skill published in the GitHub repository OSideMedia/higgsfield-ai-prompt-skill (531 stars, last pushed 20d ago), licensed MIT. It adds 165 tokens to every session and 8,775 once invoked, about $0.0008 per session on Opus 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 2026-08-30.
Other skills, from other repositories
vibe-creating-prompt
Judges whether a user's input suits the Vibe Creating style of video-prompt writing, and when it does, distills single-scene prompts, multi-shot descriptions, emotional imagery, or mixed input into prompts that are easier for a video model to generate from — while preserving any user-specified dialogue, voiceover…
seedance-cinematic-film
Write photorealistic live-action cinematic Seedance 2.0 prompts for Higgsfield, built on five grounding pillars that stop AI drift and floaty motion. Use for "cinematic film prompt", "shot like a movie", realistic body movement, grounded motion, restrained emotional close-ups, driving scenes, fight choreography…
director-cinematography
Virtual director and cinematographer for AI-generated short videos using Higgsfield. Takes a script, scene, or creative idea and produces a shot-by-shot visual direction plan with ready-to-paste Higgsfield prompts — camera presets, lenses, lighting, color grading, and motion directives for every shot. Channels 11…
seedance-prompt-builder
Write production-ready Seedance 2.0 prompts (image, video, and motion/edit) for Higgsfield. Use this skill whenever the user wants a single Seedance prompt rather than a full shotlist — e.g. "make a Seedance prompt for X", "write me a video prompt", "restyle this clip", "turn my photo into a character sheet"…
down-skilling
Distill Opus-level reasoning into optimized instructions for Haiku 4.5 (and Sonnet). Generates explicit, procedural prompts with n-shot examples that maximize smaller model performance on a given task. Use when user says "down-skill", "distill for Haiku", "optimize for Haiku", "make this work on Haiku", "generate…
reviewing-ai-papers
Analyzes an AI/ML publication — paper, preprint, article, technical blog post — and extracts what an enterprise AI engineer should do about it. Use when someone supplies a URL or document on RAG, embeddings, fine-tuning, prompt engineering, agents, or LLM deployment and asks "review this paper", "what do you make of…