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.
npx skills add dcc-mcp/dcc-mcp-blender --skill dcc-blender-assembly-pitfallsgit clone --depth 1 https://github.com/dcc-mcp/dcc-mcp-blenderWrote 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/dcc-mcp/dcc-mcp-blender/dcc-blender-assembly-pitfalls)<a href="https://agentmods.dev/skills/dcc-mcp/dcc-mcp-blender/dcc-blender-assembly-pitfalls"><img src="https://agentmods.dev/badge/skills/dcc-mcp/dcc-mcp-blender/dcc-blender-assembly-pitfalls/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/dcc-mcp/dcc-mcp-blender/dcc-blender-assembly-pitfalls"><img src="https://agentmods.dev/badge/skills/dcc-mcp/dcc-mcp-blender/dcc-blender-assembly-pitfalls.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 | $0.00153 | $0.06383 |
| Opus 5.5 | $0.00061 | $0.02553 |
| Sonnet 5.5 | $0.00031 | $0.01277 |
| Haiku 4.5 | $0.00015 | $0.00638 |
Grade A, and why
dcc-blender-assembly-pitfalls 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 18d 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 — 502 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Blender assembly pitfalls
This is a failure-mode skill, not a tool catalog. Every number below was
measured on Blender 4.5.13 LTS (daeeeca98fb0) in a headless
blender --background --factory-startup session. If a claim here cannot be
reproduced, the claim is wrong - fix the skill, do not work around the symptom.
Read it when you are about to create, size, orient, or join geometry. Skip it for read-only scene inspection.
The failure this skill prevents always has the same shape: the code runs, no exception is raised, the viewport looks plausible, and the model is wrong. Nothing in Blender warns you. So the defence is not "be careful" - it is the Phase 1 connection map and the Phase 3 checklist at the end.
Pitfall 1 - size is a full edge, radius is a half edge, and scale= halves your intent
The trap
primitive_cube_add(size=N) makes a cube N across, centred on the origin,
so its vertices sit at +/- N/2. Cylinders, spheres, and cones take radius,
which is half the extent. A request for "0.4" therefore produces a 0.4
half-extent for a cube and a 0.4 full extent for a sphere. Same number,
2x different object.
Measured
| Call | World extent X | What the request actually gave |
|---|---|---|
primitive_cube_add(size=1.0) |
1.0 | vertices at +/-0.5, not +/-1.0 |
same cube, then scale = (0.4, 0.4, 0.4) |
0.4 | half-extent 0.2, not 0.4 |
primitive_cube_add(size=0.8) |
0.8 | half-extent 0.4 - the intended result |
primitive_uv_sphere_add(radius=0.5) |
1.0 | radius is a half extent |
primitive_cylinder_add(radius=0.5, depth=1.0) |
1.0 | radius is a half extent |
primitive_cone_add(radius1=0.5, radius2=0, depth=1.0) |
1.0 | radius is a half extent |
primitive_cube_add(size=0.5) |
0.5 | size is a full extent |
The scale = 0.4 row is the one that costs hours: cube.dimensions afterwards
reads (0.4, 0.4, 0.4), so a model that verifies its work by reading
dimensions sees the number it asked for and reports success - while the part
is half the size it needs to be. dimensions is the full extent; a mating
calculation needs the half extent.
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.
- 18d ago First seen · 502 lines · 153 tokens per session scan A c5266d9e41c0
dcc-blender-assembly-pitfalls is a skill published in the GitHub repository dcc-mcp/dcc-mcp-blender (47 stars, last pushed yesterday), licensed MIT. It adds 153 tokens to every session and 6,383 once invoked, about $0.0006 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 2026-09-28.
Other skills, from other repositories
blender-reference-modeling
Build, compare, review, or repair a Blender model against an actual supplied image, photograph, concept, or visual reference using adaptive decomposition, measurable proportions, evidence scoring, and bounded repair. Use only when the request explicitly concerns model creation or quality against visual form criteria…
blender-retopology
Improve Meshy, Tripo or other existing Blender meshes with deliberate editable topology, source-preserving cleanup, reduction, remeshing or reference-derived reconstruction, followed by geometry and appearance checks. Use for dense generated meshes, human-style edge flow, hard-surface rebuilding or character…
blender-bridge
Operate Blender Agent Bridge through its stable five-tool MCP gateway for scene inspection, modeling, materials, rigging, animation, simulation, rendering, external assets, trusted scripts, and preview decisions. Use whenever an agent must work in a live Blender scene through Blender Bridge, especially when only…
blender-asset-validation
Inspect and validate Blender assets technically and visually. Use for .blend, .glb, .gltf, .fbx, or .obj quality checks; topology and export review; evaluated triangle/material/hierarchy metrics; standardized multiview renders; fresh-import verification; or evidence-backed review of an agent-generated mesh.
blender-modeling-workflow
Build or substantially refine reproducible Blender models through Python and the Blender CLI. Use for user requests to create meshes, props, hard-surface assets, stylized objects, assemblies, procedural geometry, game-ready GLB assets, or to iterate on a scripted Blender model from visual feedback.
dcc-cua
A routing guide for controlling application interfaces through the project’s DCC-CUA system. DCC applications are digital-content tools such as Maya, Blender, and Houdini.