Agent skill

Minimax H3 Multishot Planner

by unknowlei in unknowlei/minimax-h3-opencode-skills

Non-skippable planning-only MiniMax H3 subskill used by minimax-h3-creative-director before final prompt formatting.

MITAuto-check passedMedia & Creative

Install Minimax H3 Multishot Planner

skills CLI
$ npx skills add unknowlei/minimax-h3-opencode-skills --skill minimax-h3-multishot-planner -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install unknowlei/minimax-h3-opencode-skills minimax-h3-multishot-planner --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/unknowlei/minimax-h3-opencode-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/minimax-h3-multishot-planner .claude/skills/minimax-h3-multishot-planner && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
minimax-h3-multishot-planner
GitHub stars
122
Token cost
~2.2k tokens
SKILL.md length
1,082 words
Files
3 (incl. references)
Skills in repo
6
Repo updated
First seen
Licence
MIT

At a glance

Non-skippable planning-only MiniMax H3 subskill used by minimax-h3-creative-director before final prompt formatting.

  • Works in 6 steps: Do not load any final H3 specialist. → Do not draft an H3 prompt. → Ask or confirm shot count. → …
  • Explicitly requests multiple shots
  • SKILL.md covers Execution Is Mandatory After…, Invocation Contract, Choice Source and Question Payload Rules, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Minimax H3 Multishot Planner is an agent skill from unknowlei/minimax-h3-opencode-skills. Non-skippable planning-only MiniMax H3 subskill used by minimax-h3-creative-director before final prompt formatting. Invoke when the user explicitly requests multiple shots, scenes, cuts, montage, or shot-by-shot design, or when a video of at least 10 seconds has no declared single-shot/multishot preference and the user accepts the director's multishot offer. After loading, it must ask shot count and then question the user about every shot in playback order before any final H3 formatting skill may run.

Its SKILL.md is about 2.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `agents/openai.yaml` and `references/shot-choice-library.md`).

It sits in Media & Creative. It works with MiniMax. The repository describes itself as: OpenCode skill suite for MiniMax H3 directing, routing, multishot planning, prompt generation, and review. The licence is MIT.

When your agent uses it

  • Explicitly requests multiple shots
  • Shot-by-shot design
  • A video of at least 10 seconds has no declared single-shot/multishot preference and the user accepts the directors multishot offer

Example prompts

  • “/minimax-h3-multishot-planner”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. Do not load any final H3 specialist.
  2. Do not draft an H3 prompt.
  3. Ask or confirm shot count.
  4. Question the user about Shot 1, confirm it, then repeat for Shot 2 and every remaining shot.
  5. Build and confirm the complete multishot_plan.
  6. Return control to the director only after final confirmation.

What it can do on your machine

Read from SKILL.md and the folder at commit 2e2096f. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Minimax H3 Multishot Planner loads about 2.2k tokens when it runs, and up to ~3.4k if it reads all its reference files. Until then it costs about 134 tokens; SKILL.md has 1,082 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~134
When it runs · the whole SKILL.md, loaded when a task matches
~2.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~3.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from unknowlei/minimax-h3-opencode-skills at commit 2e2096f, republished under its MIT licence (© unknowlei). 1,082 words, ~2,166 tokens.

Download SKILL.mdSave it as .claude/skills/minimax-h3-multishot-planner/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
minimax-h3-multishot-planner
description
Non-skippable planning-only MiniMax H3 subskill used by minimax-h3-creative-director before final prompt formatting. Invoke when the user explicitly requests multiple shots, scenes, cuts, montage, or shot-by-shot design, or when a video of at least 10 seconds has no declared single-shot/multishot preference and the user accepts the director's multishot offer. After loading, it must ask shot count and then question the user about every shot in playback order before any final H3 formatting skill may run.

MiniMax H3 Multishot Planner

Plan a confirmed multishot video before the text, keyframe, or full-reference specialist writes the executable H3 prompt. Produce a structured shot plan, not the final H3 prompt.

Execution Is Mandatory After Loading

The skill tool only loads these instructions. Loading this Skill is not execution and is never a successful handoff by itself.

After this Skill is loaded:

  1. Do not load any final H3 specialist.
  2. Do not draft an H3 prompt.
  3. Ask or confirm shot count.
  4. Question the user about Shot 1, confirm it, then repeat for Shot 2 and every remaining shot.
  5. Build and confirm the complete multishot_plan.
  6. Return control to the director only after final confirmation.

No final specialist may run while any planned shot remains unasked or unconfirmed.

Invocation Contract

Enter only through minimax-h3-creative-director:

  • Enter immediately when the user explicitly requests multiple shots, scenes, cuts, a montage, a sequence, or shot-by-shot planning.
  • When effective duration is at least 10 seconds and the user has not chosen single-shot or multishot, let the director ask a binary multishot question. Enter only after the user accepts.
  • Do not enter when the user explicitly requests one continuous shot, one take, no cuts, or an equivalent single-shot treatment.
  • Do not enter merely because a prompt contains several actions. A continuous action arc can remain one shot.
  • Do not choose T2VA, I2VA, FL2VA, L2VA, or Ref2VA here.

If duration is unknown, return control to the director to resolve duration before applying the 10-second gate.

Choice Source

Consult references/shot-choice-library.md for shot vocabulary and option design. Do not treat it as an official closed vocabulary or insert every available term.

Question Payload Rules

OpenCode does not impose a three-question workflow limit. Use the number of questions required by each state below.

  • A strict yes/no question may contain exactly two options.
  • Every other choice question must contain at least five materially different, feasible options.
  • Put the strongest context-specific recommendation first and append (Recommended) to its label.
  • Explain the creative consequence of every option in one concise sentence.
  • Keep every option label within five words; place detail in its description.
  • Do not add Other; OpenCode supplies custom input automatically.
  • Do not pad a list with near-duplicates. If five meaningful choices cannot be produced, ask an open-ended question instead of a choice question.
  • Generate content and performance choices from the user's actual subject, assets, genre, duration, and desired outcome. Never present generic placeholders.
  • Use multiple selection only when choices can genuinely coexist.

Before every question call, count the options for every question. If a non-binary question has fewer than five options, the payload is invalid: expand it or make that question open-ended before calling the tool.

If the user explicitly prohibits questions, synthesize the complete plan, disclose material assumptions, and ask one binary final confirmation. Otherwise, all states below are mandatory.

State Machine

State 1: Shot count

This is the first action after loading unless the user already supplied an exact count.

  • Ask exactly one shot-count question.
  • Provide at least five feasible counts, normally 2, 3, 4, 5, and 6 shots.
  • Recommend according to duration and action density.
  • If an exact count was already supplied, record it and advance directly to State 2 without asking the same question again.

Starting heuristics, not hard limits:

  • 4-9 seconds: usually 2-3 shots only when multishot was explicitly requested.
  • 10-12 seconds: usually 2-4 shots.
  • 13-15 seconds: usually 3-5 shots.

If the chosen count is physically infeasible, ask the strict yes/no question: "是否改用推荐镜头数?"

State 2: Global timing and continuity

Ask one batch with up to three questions when these decisions remain unknown:

  1. Story or information progression
  2. Timing distribution and edit rhythm
  3. Continuity strategy across cuts

Every choice question in this batch requires at least five options. Allocate provisional start/end times after receiving the answers. Do not force equal durations when action complexity differs.

Show full SKILL.md (439 more words)Show less
State 3: Plan each shot sequentially

Maintain current_shot, starting at 1. Never combine different shots in one question payload and never jump ahead.

For the current shot, make one question call containing the following six questions unless a field was already explicitly fixed by the user:

  1. Shot content and narrative function: at least five scene-specific alternatives.
  2. Shot size and composition: at least five suitable framing/composition alternatives.
  3. Camera behavior: at least five complete treatments combining movement type, target, speed, amplitude, and compositional purpose.
  4. Action, performance, and expression: at least five scene-specific visible action arcs.
  5. Opening or entry transition: at least five opening treatments for Shot 1; at least five cut/transition treatments for later shots.
  6. Sound focus: at least five relevant combinations of ambience, Foley, non-verbal vocal sound, dialogue/voice, diegetic sound, and audience-only music.

This six-question per-shot batch is intentional. Do not reduce it to three merely because earlier general H3 questions used three-question batches. Omit only a field the user has already explicitly and unambiguously fixed; do not silently infer a missing field to avoid asking.

After receiving the batch answers:

  1. Show a concise Chinese summary of the current shot, including provisional timing and active reference roles.
  2. Ask the strict yes/no question: "是否确认 Shot N?"
  3. If no, ask an open-ended correction question and revise the same shot.
  4. If yes, increment current_shot and repeat State 3 for the next shot.

Do not enter State 4 until every shot from 1 through shot_count is confirmed.

State 4: Build continuity

After all shots are confirmed:

  • Allocate exact start and end times; Shot 1 starts at 0.00 seconds and later start times increase strictly.
  • Record the entering and exiting state of every shot.
  • Preserve identity, clothing, props, scene geography, lighting direction, object state, visible text, dialogue continuity, and audio continuity unless a change is intentional.
  • Assign active reference labels and responsibilities per shot when assets exist.
  • Identify J-cuts, L-cuts, match cuts, or sound bridges only when their cross-shot relationship is explicit.
  • Reject a multishot plan that also claims to be one continuous take.
State 5: Final confirmation and handoff

Present the complete plan in Chinese using the schema below. Ask the strict yes/no question: "是否接受这份多镜头方案?"

  • If no, ask an open-ended question identifying the shot and desired change, revise it, then reconfirm.
  • If yes, mark multishot_plan_status: confirmed and return the plan to minimax-h3-creative-director.

Only multishot_plan_status: confirmed releases the director's hard blocker and permits one final prompt-format specialist to load.

Output Schema

text
multishot_plan_status: confirmed
multishot_plan:
  total_duration: ...
  shot_count: ...
  edit_rhythm: ...
  continuity_strategy: ...
  shots:
    - shot: 1
      start: 0.00
      end: ...
      narrative_function: ...
      content: ...
      shot_size_and_composition: ...
      subject_action_and_expression: ...
      camera_movement_amplitude_speed: ...
      entry_transition: opening
      sound_focus: ...
      active_references: ...
      entry_state: ...
      exit_state: ...
    - shot: 2
      ...
  continuity_ledger:
    identity: ...
    wardrobe_and_props: ...
    space_and_lighting: ...
    object_and_text_state: ...
    dialogue_and_audio: ...

The final formatting specialist converts this confirmed plan to the official H3 base three-field or Ref2VA six-section prompt without dropping or re-asking confirmed decisions.

© unknowlei, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (references) in skills/minimax-h3-multishot-planner of unknowlei/minimax-h3-opencode-skills.

  • SKILL.md
  • agents/openai.yaml
  • references/shot-choice-library.md

Open the folder on GitHubat commit 2e2096f

Compare with similar skills

Minimax H3 Multishot Planner next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Minimax H3 Multishot Planner compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Minimax H3 Multishot Planner this skillunknowlei/minimax-h3-opencode-skills122—~2.2kAutomated safety check: PassMIT
AI Image Generation and Editingzhayujie/CowAgent47k—~1.3kAutomated safety check: PassMIT
Gif Sticker MakerHHU3637kr/skills1453 repos~1.3kAutomated safety check: PassMIT
Minimax H3 Video Reversegnipbao/minimax-h3-video-reverse-skill230—~2.2kAutomated safety check: PassMIT
H3 Cinematic Directorjtydhr88/ComfyTV1.1k—~2.7kAutomated safety check: PassMIT
Novel Storyboardeternityspring/shuohao-skills4.3k—~1.8kAutomated safety check: NotesApache-2.0

Similar skills

  • Generates or edits images from text prompts through a Python script that picks an image backend based on which API keys are configured.

    47k GitHub stars~1.3k tokensUpdated yesterday
    Media & CreativeAuto-check passed
  • Gif Sticker Maker

    HHU3637kr/skills

    Convert photos (people, pets, objects, logos) into 4 animated GIF stickers with captions.

    145 GitHub starsUsed in 3 repos~1.3k tokens
    Media & CreativeAuto-check passed
  • Minimax H3 Video Reverse

    gnipbao/minimax-h3-video-reverse-skill

    先复述参考视频内容与叙事逻辑,和用户确认理解后,再反推 MiniMax H3 文生视频或逐镜图生视频提示词。Use when the user asks to reverse-engineer a video into prompts, reconstruct shots, extract first/end-frame prompts, adapt a reference to H3, or…

    230 GitHub stars~2.2k tokensUpdated 8 days ago
    Media & CreativeAuto-check passed
  • H3 Cinematic Director

    jtydhr88/ComfyTV

    Convert approved scripts, shot briefs, storyboards, keyframes, character sheets, scene assets, prop sheets, and sound briefs into director-level storyboards and production-ready MiniMax H3 prompts.

    1.1k GitHub stars~2.7k tokensUpdated 4 days ago
    Media & CreativeAuto-check passed
  • Novel Storyboard

    eternityspring/shuohao-skills

    给 AI 短剧出分镜:三层结构——段(一次视频生成,≤15 秒)→ 分镜(段内 2–5 秒的剪切,认领剧本节拍) → 分镜图(每切一张关键帧:主分镜图钉 0.00 秒,子分镜图钉各自切点)。

    4.3k GitHub stars~1.8k tokensUpdated 3 days ago
    Media & CreativeAuto-check: notes
  • Minimax H3 Colab

    killkli/minimax-h3-colab-skill

    Create short MiniMax H3 reference-to-video clips from one or more local images through Google Colab CLI.

    110 GitHub stars~743 tokensUpdated 14 days ago
    Media & CreativeAuto-check passed

More from unknowlei/minimax-h3-opencode-skills

  • Minimax H3 Keyframe Video Prompt

    unknowlei/minimax-h3-opencode-skills

    Narrow downstream MiniMax H3 specialist for pure I2VA, FL2VA, and L2VA boundary-frame prompts.

    122 GitHub stars~2.4k tokensUpdated 2 mo ago
    Auto-check passed
  • Minimax H3 Reference Video Prompt

    unknowlei/minimax-h3-opencode-skills

    Default downstream MiniMax H3 specialist for every image-based request unless the user explicitly declares boundary-only first/last frames with no reusable reference role.

    122 GitHub stars~3k tokensUpdated 2 mo ago
    Auto-check passed
  • Minimax H3 Text Video Prompt

    unknowlei/minimax-h3-opencode-skills

    Downstream MiniMax H3 specialist for professional text-to-video (T2VA) prompts using the official three-field format.

    122 GitHub stars~2.3k tokensUpdated 2 mo ago
    Auto-check passed
  • Minimax H3 Creative Director

    unknowlei/minimax-h3-opencode-skills

    Primary mandatory entrypoint for every MiniMax H3 video-generation request.

    122 GitHub stars~2.8k tokensUpdated 2 mo ago
    Auto-check passed
  • Minimax H3 Prompt Reviewer

    unknowlei/minimax-h3-opencode-skills

    Downstream MiniMax H3 specialist that audits, repairs, and rewrites T2VA, I2VA, FL2VA, L2VA, and full-reference prompts into an official structured format.

    122 GitHub stars~2.1k tokensUpdated 2 mo ago
    Auto-check passed

Works with

Questions about Minimax H3 Multishot Planner

What does Minimax H3 Multishot Planner do?

Non-skippable planning-only MiniMax H3 subskill used by minimax-h3-creative-director before final prompt formatting. Minimax H3 Multishot Planner is an agent skill from unknowlei/minimax-h3-opencode-skills. Non-skippable planning-only MiniMax H3 subskill used by minimax-h3-creative-director before final prompt formatting.

When should I use Minimax H3 Multishot Planner?

Minimax H3 Multishot Planner fits situations like: explicitly requests multiple shots; shot-by-shot design; A video of at least 10 seconds has no declared single-shot/multishot preference and the user accepts the directors multishot offer.

How do I install Minimax H3 Multishot Planner in Claude Code?

Run `npx skills add unknowlei/minimax-h3-opencode-skills --skill minimax-h3-multishot-planner -a claude-code`. Or copy the skill folder (skills/minimax-h3-multishot-planner in unknowlei/minimax-h3-opencode-skills) into .claude/skills/minimax-h3-multishot-planner in your project. Claude Code loads it when a task matches its description.

How do I install Minimax H3 Multishot Planner in Codex?

Run `npx skills add unknowlei/minimax-h3-opencode-skills --skill minimax-h3-multishot-planner -a codex`. Or copy the skill folder (skills/minimax-h3-multishot-planner in unknowlei/minimax-h3-opencode-skills) into .agents/skills/minimax-h3-multishot-planner in your project. Codex loads it when a task matches its description.

Can I use Minimax H3 Multishot Planner in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add unknowlei/minimax-h3-opencode-skills --skill minimax-h3-multishot-planner -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/minimax-h3-multishot-planner, .gemini/skills/minimax-h3-multishot-planner, .github/skills/minimax-h3-multishot-planner and .opencode/skills/minimax-h3-multishot-planner in your project.

What does Minimax H3 Multishot Planner need to run?

SKILL.md names no scripts, command-line tools or credentials: Minimax H3 Multishot Planner is instructions for the agent only.

Does Minimax H3 Multishot Planner access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Minimax H3 Multishot Planner safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Minimax H3 Multishot Planner use?

Minimax H3 Multishot Planner is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Minimax H3 Multishot Planner use?

About 2.2k tokens (SKILL.md is roughly 8.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.2k tokens, read only when the agent opens those files.

What are the alternatives to Minimax H3 Multishot Planner?

Skills that share tags, products or a category with Minimax H3 Multishot Planner: AI Image Generation and Editing (zhayujie/CowAgent, 47k stars), Gif Sticker Maker (HHU3637kr/skills, 145 stars), Minimax H3 Video Reverse (gnipbao/minimax-h3-video-reverse-skill, 230 stars) and H3 Cinematic Director (jtydhr88/ComfyTV, 1.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Minimax H3 Multishot Planner?

unknowlei (a GitHub user) maintains it in unknowlei/minimax-h3-opencode-skills, which has 122 GitHub stars. The repository holds 6 skills in this directory. The repository was last updated on August 9, 2026.

Source: unknowlei/minimax-h3-opencode-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.