Animsequence logo

Animsequence

Community
kevinpbuckley
animsequence

Create and edit AnimSequence keyframes and tracks — add/remove keyframes, set bone tracks, query sequence info, and bake edits (AnimSequenceService). Use when the user asks to create a new animation sequence, add keyframes/tracks, or edit raw animation data. For constraint-aware bone posing, also load animation-editing.

Overview

Publisherkevinpbuckley
RepositoryVibeUE
Skill nameanimsequence
Stars
735
Forks
161
Bundled files
7
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 7 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by kevinpbuckley on GitHub. Read the source before you install it.

Installation

Install the Animsequence AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/kevinpbuckley/VibeUE/tree/master/Content/Skills/animsequence
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/kevinpbuckley/VibeUE.git /tmp/VibeUE
mkdir -p .claude/skills
cp -r /tmp/VibeUE/Content/Skills/animsequence .claude/skills/animsequence
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Animsequence in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Animsequence on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Animsequence is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

🧠 Brains complement: IF an unreal-engine-skills-manager tool (external MCP) exists in this session, call it with {action: "load", skill: "animation-system"} for UE domain knowledge on this topic — correct APIs, architecture, best practices — and treat it as the rubric for any review / "best practices" question. If no such tool is available (e.g. running under Claude Code or Codex without that MCP), skip this line entirely and proceed with this skill alone — do NOT attempt the call.

Animation Sequence & Editing Skill

This skill covers the inspect → profile → preview → validate → bake workflow for safe animation bone edits with correct bone-space handling and constraint validation.

Loading skills: skills load via the engine's AgentSkillToolset (ListSkills / GetSkills) — there is no vibeue-skills-manager tool. Run the VibeUE services below with execute_python_code (unreal.AnimSequenceService.<method>() / unreal.SkeletonService.<method>()).

Related Skills:

  • skeleton - For modifying skeleton bone transforms/hierarchy and retargeting modes
  • animation-blueprint - For state machines and AnimGraph navigation

Workflow: Create skeleton profile → Learn constraints → Preview bone rotations → Validate pose → Bake to keyframes

🛡️ SAFE DISCOVERY: Always use VibeUE service methods

DO NOT use unreal.load_asset() in loops - causes memory crashes!

USE THESE SAFE METHODS:

  • unreal.SkeletonService.list_skeletons(search_path) - Find skeletons
  • unreal.AnimSequenceService.find_animations_for_skeleton(skeleton_path) - Find animations
  • unreal.AnimSequenceService.list_anim_sequences(search_path, skeleton_filter) - List all animations

These methods query asset metadata WITHOUT loading assets into memory.

  • animation-blueprint - For state machines and AnimGraph navigation

Workflow: Create skeleton profile → Learn constraints → Preview bone rotations → Validate pose → Bake to keyframes

⚠️ Critical Rules

1. Always Specify Bone Space

All bone rotations must specify the coordinate space. Default to "local" for user intent.

SpaceDescriptionUse For
"local"Relative to parent boneMost edits (default)
"component"Relative to skeletal mesh rootCross-bone coordination
"world"World coordinatesRarely used for animations
python
# WRONG: Assuming space or not specifying
unreal.AnimSequenceService.apply_bone_rotation(path, "arm", rot, ...)

# CORRECT: Always specify space explicitly
unreal.AnimSequenceService.preview_bone_rotation(
    path, "upperarm_r", unreal.Rotator(0, 30, 0), "local", 0
)

2. Use Preview → Validate → Bake Workflow

Never apply edits directly without validation. Use the preview workflow:

python
import unreal

anim_path = "/Game/Anims/AS_Idle"

# Step 1: PREVIEW the edit
result = unreal.AnimSequenceService.preview_bone_rotation(
    anim_path, "upperarm_r", unreal.Rotator(0, 45, 0), "local", 0
)

# Step 2: VALIDATE against constraints
validation = unreal.AnimSequenceService.validate_pose(anim_path, True)  # Use learned constraints
if not validation.is_valid:
    for msg in validation.violation_messages:
        print(f"⚠️ {msg}")
    # Option A: Cancel and adjust
    unreal.AnimSequenceService.cancel_preview(anim_path)
    # Option B: Accept clamped values and continue

# Step 3: BAKE if valid
if validation.is_valid:
    result = unreal.AnimSequenceService.bake_preview_to_keyframes(
        anim_path, 0, -1, "cubic"
    )
    print(f"✓ Baked frames {result.start_frame} to {result.end_frame}")

3. Build Skeleton Profile Before Editing

Create a skeleton profile to get hierarchy, constraints, and learned ranges:

python
# WRONG: Editing without understanding the skeleton
unreal.AnimSequenceService.apply_bone_rotation(...)

# CORRECT: Build profile first
profile = unreal.SkeletonService.create_skeleton_profile("/Game/SK_Mannequin")
print(f"Skeleton has {profile.bone_count} bones")

# Check if constraints are available
if not profile.has_learned_constraints:
    # Learn from existing animations
    constraints = unreal.SkeletonService.learn_from_animations("/Game/SK_Mannequin", 50, 10)
    print(f"Learned from {constraints.animation_count} animations")

4. Use Quaternions Internally, Euler for Intent

User intent is expressed in Euler angles (degrees), but always use quaternions internally to avoid gimbal lock:

python
# User intent: "rotate forearm 30 degrees"
rotation_delta = unreal.Rotator(0, 30, 0)  # Euler (Roll, Pitch, Yaw)

# The service converts to quaternion internally
result = unreal.AnimSequenceService.preview_bone_rotation(
    anim_path, "lowerarm_r", rotation_delta, "local", 0
)

# To inspect quaternion values:
euler = unreal.AnimSequenceService.quat_to_euler(some_quat)

COMMON_MISTAKES

Wrong: Applying rotations without space

python
# WRONG - space is ambiguous
transform = unreal.Transform(rotation=some_rotation)
# This may be interpreted incorrectly

# CORRECT - always specify space
result = unreal.AnimSequenceService.preview_bone_rotation(
    anim_path, bone, rotation, "local", frame
)

Wrong: Editing without validation

python
# WRONG - direct edit without checking constraints
unreal.AnimSequenceService.apply_bone_rotation(
    path, "lowerarm_r", unreal.Rotator(0, 180, 0),  # Impossible elbow bend!
    "local", 0, -1, True
)

# CORRECT - preview → validate → bake
result = unreal.AnimSequenceService.preview_bone_rotation(...)
validation = unreal.AnimSequenceService.validate_pose(path, True)
if validation.is_valid:
    unreal.AnimSequenceService.bake_preview_to_keyframes(...)

Wrong: Guessing bone axis orientations

python
# WRONG - assuming pitch raises arm
rotation = unreal.Rotator(0, 90, 0)  # May not do what you expect!

# CORRECT - analyze reference pose first
ref_pose = unreal.AnimSequenceService.get_reference_pose(skeleton_path)
for bp in ref_pose:
    if bp.bone_name == "upperarm_r":
        euler = unreal.AnimSequenceService.quat_to_euler(bp.transform.rotation)
        print(f"Reference: Roll={euler.roll}, Pitch={euler.pitch}, Yaw={euler.yaw}")
        # Now you know the baseline to add deltas to

Wrong: Not building skeleton profile

python
# WRONG - editing without profile
unreal.AnimSequenceService.preview_bone_rotation(...)  # No constraints!

# CORRECT - build profile first
unreal.SkeletonService.create_skeleton_profile(skeleton_path)
# Now constraints are available for validation

⚠️ CRITICAL: Loading assets in loops causes crashes

python
# WRONG - Loading assets causes memory access violations (0xC0000005)
asset_subsystem = unreal.get_editor_subsystem(unreal.EditorAssetSubsystem)
assets = asset_subsystem.list_assets("/Game", recursive=True)
for asset_path in assets:
    anim = unreal.load_asset(asset_path)  # ❌ CRASHES!
    if anim and hasattr(anim, 'get_skeleton'):
        skeleton = anim.get_skeleton()  # Memory violation

# CORRECT - Use Asset Registry API (no loading required)
asset_registry = unreal.AssetRegistryHelpers.get_asset_registry()
filter = unreal.ARFilter(
    class_paths=[unreal.TopLevelAssetPath("/Script/Engine", "AnimSequence")],
    recursive_paths=True,
    package_paths=["/Game"]
)
anim_assets = asset_registry.get_assets(filter)
for asset_data in anim_assets:
    # Access metadata without loading
    anim_name = asset_data.asset_name
    anim_path = asset_data.object_path
    # Get skeleton path from asset registry tags
    skeleton_tag = asset_data.get_tag_value("Skeleton")

Why this crashes:

  • unreal.load_asset() loads full asset into memory
  • Loading hundreds of animations in a loop exhausts memory
  • Unreal's garbage collection can't keep up
  • Results in exception code 0xC0000005 (access violation)

Safe alternatives:

  1. USE VibeUE's FindAnimationsForSkeleton() - BEST OPTION
  2. Use Asset Registry API to query metadata without loading
  3. Load one asset at a time with explicit unload
  4. Filter by path before loading (limit scope)

BEST: Use VibeUE AnimSequenceService methods

python
# CORRECT - VibeUE safe discovery (no loading)
import unreal

skeleton_path = "/Game/Characters/SK_Mannequin"

# Get all animations for a skeleton - no loading!
anims = unreal.AnimSequenceService.find_animations_for_skeleton(skeleton_path)

for anim_info in anims[:10]:
    print(f"{anim_info.anim_name}")
    print(f"  Duration: {anim_info.duration:.2f}s")
    print(f"  Frames: {anim_info.frame_count}")
    print(f"  Path: {anim_info.anim_path}")

Other safe VibeUE methods:

  • list_anim_sequences(search_path, skeleton_filter) - List all in path
  • get_anim_sequence_info(anim_path) - Get single anim info
  • search_animations(name_pattern, search_path) - Find by name pattern

IMPORTANT: Asset Path Format

All anim_path parameters require the FULL asset path (package_name), NOT the folder path (package_path).

When using AssetDiscoveryService.search_assets(), use package_name NOT package_path:

python
import unreal

# Search for an animation
results = unreal.AssetDiscoveryService.search_assets("Run", "AnimSequence")
if results:
    asset = results[0]
    
    # CORRECT: Use package_name (full asset path)
    anim_path = str(asset.package_name)  # e.g., "/Game/Animations/Run/AS_Run_Forward"
    
    # WRONG: Do NOT use package_path (folder only)
    # folder = asset.package_path  # e.g., "/Game/Animations/Run" - This will FAIL!
    
    # Now you can use the anim_path with AnimSequenceService
    info = unreal.AnimSequenceService.get_anim_sequence_info(anim_path)

Sub-docs available

Load these sibling docs for deeper coverage of specific topics:

  • api-reference.md - Full Python API reference (SkeletonService profile/constraint methods, AnimSequenceService preview/edit/pose/visual capture, creation/helper/query methods, key structs). Also contains the secondary skill manifest for the merged "Animation Sequence Skill" content set.
  • data-structures.md - Key data structures (FSkeletonProfile, FBoneConstraint, FLearnedBoneRange, FAnimationEditResult, FAnimationPoseCaptureResult, FBoneDelta) and full DTO reference (AnimSequenceInfo, BoneTrackData, AnimKeyframe, BonePose, AnimCurveInfo, CurveKeyframe, AnimNotifyInfo, SyncMarkerInfo).
  • workflows.md - Editing workflows (inspect, preview/validate single bone, multi-bone atomic, manual constraints, learn from animations, copy/mirror pose, retarget preview) and creation workflows (from reference pose, from keyframe data, simple keyframes, rotation keyframes, wave animation example).
  • patterns.md - Common patterns (listing animations, getting pose at time, adding curves, managing/configuring notifies) and the CRITICAL RULES checklist for creation/notify operations and asset paths.
  • method-categories.md - Categorised table of every AnimSequenceService method by domain (Discovery, Creation, Helpers, Properties, Bone Tracks, Poses, Curves, Notifies, Notify Tracks, Sync Markers, Root Motion, Additive, Compression, Export, Editor).
  • bone-rotation.md - Bone Rotation Axis Discovery: why bone local coordinate systems are non-intuitive and the required discovery workflow (reference pose values → analyze existing animations → incremental axis testing) before writing rotation animation code.

Sample scripts (run via execute_python_code)

  • scripts/inspect_sequence.txt — safely inspect AnimSequences for a skeleton (metadata, no asset loading).

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Animsequence AI skill do?

Create and edit AnimSequence keyframes and tracks — add/remove keyframes, set bone tracks, query sequence info, and bake edits (AnimSequenceService). Use when the user asks to create a new animation sequence, add keyframes/tracks, or edit raw animation data. For constraint-aware bone posing, also load animation-editing.

Why use Animsequence on TypingMind?

Because you install it once and use it with any model. Animsequence is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Animsequence in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/kevinpbuckley/VibeUE/tree/master/Content/Skills/animsequence. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Animsequence?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Animsequence?

As many as you like. As long as a model supports skills, you can use Animsequence with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Animsequence AI skill free?

Yes. It is published on GitHub by kevinpbuckley under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇