setup-autodesk-fusion
Configure and diagnose Fusion interoperability, transport selection, scoped profiles and qualification. Use for setup or connection problems, not design edits.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-11
- Stars
- ★ 21
- Forks
- 15
01 前言 Front Matter
- name
- setup-autodesk-fusion
- description
- Configure and diagnose Fusion interoperability, transport selection, scoped profiles and qualification. Use for setup or connection problems, not design edits.
02 资源文件 2
- 参考文档agents/openai.yaml207 B
- 参考文档references/scoring-rubric.md5.8 KB
03 SKILL.md 正文
Call fusion_connection_status and fusion_capabilities_list first. Distinguish synthetic fixture, managed and assisted modes. Fixture output is not evidence that Autodesk is installed.
Read setup and resolve the installed plugin root. Use its bundled CLI without startup npm. Discover the explicitly configured native endpoint and enroll the actual Python tool/schema hash; never guess tools, scan ports or silently switch transports. For the typed add-in, install the assembled package and follow the add-in guide.
For additional required native arguments, use native-enroll --fixed-arguments-file with a trusted absolute JSON-object file and explicitly reviewed values. Do not infer defaults, reuse an old mapping's values or override the script argument. An optional --url is saved as the selected endpoint. Enrollment rejects changed profiles and never executes the native tool or establishes qualification.
Use a trusted private profile for the user's scope. Existing valid grants cover routine work; changed target, billing, publication or manufacturing risk needs matching authority. Never edit policy to bypass a denial or invent qualification evidence.
Cloud PKCE and native Data MCP authentication are separate. Let the user complete the supported Autodesk sign-in URL; never collect passwords or copy Codex credentials. Enterprise extensions are privileged deployment code.
Run qualification on approved synthetic documents. Report the connection evidence and exact remaining configuration or licensed-environment gate.
04 同仓库更多 Skills
recommend-review-reference-architecture
Recommend a conservative shadow-only review reference architecture from valid offline evidence. Use only when the user explicitly asks for a reference-architecture recommendation or uses the bundled synthetic fixture.
normalize-review-runs
Validate and normalize declared local review-run exports into a bounded canonical summary. Use only when the user explicitly asks to normalize review runs or use the bundled synthetic fixture.
benchmark-review-configs
Deterministically score comparable imported review-run lanes with root-cause metrics and uncertainty. Use only when the user explicitly asks to benchmark normalized local review runs or use the bundled synthetic fixture.
audit-review-eval-validity
Audit whether a declared local review evaluation can support fair quality claims. Use only when the user explicitly asks to audit review-eval validity or use the bundled synthetic fixture.
verify-swiftui-parity
Verify one approved SwiftUI feature slice against its behavioral parity contract with deterministic and conditional Apple-platform checks.
port-react-native-slice
Port one explicitly approved React Native or Expo feature slice into a declared SwiftUI target while preserving its parity contract.