ChatGPT integration
Attach a deployed model and ask for its static evidence.
The ChatGPT integration opens DEEPBOM in a sandboxed browser component. It downloads the file already authorized in ChatGPT, analyzes it in that component, and returns a bounded, hash-bound result to the conversation.
Connect during review
- Enable developer mode in ChatGPT.
- Add
https://deepbom.org/mcpas a Streamable HTTP plugin endpoint. - Attach one supported deployment artifact and ask: Use DEEPBOM to inspect this file. Keep defects, cautions, and evidence gaps separate.
The endpoint is ready for private developer-mode testing. Public discovery in ChatGPT begins only after OpenAI plugin review and listing.
Ask in plain language
Use DEEPBOM to inspect this attached ONNX model.
Keep artifact defects, cautions, and evidence gaps separate.
What tensor encodings are present in this GGUF file?
Include its SHA-256 and the static evidence boundary.
Can this attached file establish runtime latency or accuracy?DEEPBOM should analyze a supported attachment for the first two prompts. For the third, it should explain that those outcomes require separate runtime or task evidence rather than inventing a static conclusion.
Choose the right execution path
| ChatGPT attachment | Convenient single-file review. Analysis executes in the ChatGPT browser sandbox and a bounded result is returned to the conversation. |
|---|---|
| Local MCP or CLI | Use for confidential or large files, package directories, ONNX external data, sharded repositories, repeated automation, and complete exports. |
| DeepBOM service | The MCP control plane serves tool metadata and validates the bounded result. It does not fetch or retain model bytes. |
Evidence boundary
DEEPBOM reports serialized-artifact observations and deterministic static derivations. It does not establish measured latency, energy, task accuracy, clinical validity, actual runtime placement, or regulatory compliance. artifact_defect, caution, and evidence_gap remain separate in the returned result.
Supported single-file inputs are TFLite, ONNX, GGUF, SafeTensors, Core ML .mlmodel, and ExecuTorch .pte/.ptd. Package and sidecar closure remain local-tool workflows.