DEEPBOM
Browser-local and CLI static evidence for deployed AI model artifacts.
我该使用它吗
质量与安全性
A
基于对工具定义和协议合规性的自动分析。
上下文开销
~6,610token 数（工具定义）
~10.7 KB典型响应大小
对注意力有显著影响（占 128k 上下文窗口的 5.16%）
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高，可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中：
{
"mcpServers": {
"deepbom": {
"command": "npx",
"args": [
"deepbom"
]
}
}
}可运行的软件包
npmdeepbom
2.0.0stdio远程端点
https://deepbom.org/mcpstreamable-http它能做什么
工具清单
工具（7）
🟢 只读🟡 写入🔴 删除⚪ 未知
🟢deepbom_analyze_file(file, analysis_depth)
Use when the user asks to inspect, summarize, or visualize an attached TFLite, ONNX, GGUF, SafeTensors, Core ML, or ExecuTorch deployment artifact. Opens a browser-local analyzer, returns a hash-bound format-neutral Model IR table and static evidence summary, and provides deterministic document views without executing model code. Do not use for training checkpoints, measured latency, task accuracy, clinical validity, regulatory compliance, or actual runtime placement.
输入模式
{
"type": "object",
"properties": {
"file": {
"$ref": "#/$defs/OpenAIFile"
},
"analysis_depth": {
"type": "string",
"enum": [
"structure",
"payload_integrity"
],
"default": "structure",
"description": "Use structure unless the user explicitly asks to inspect serialized tensor payload values. Payload integrity can require reading the complete attachment."
}
},
"required": [
"file"
],
"$defs": {
"OpenAIFile": {
"type": "object",
"properties": {
"download_url": {
"type": "string"
},
"file_id": {
"type": "string"
},
"mime_type": {
"type": "string"
},
"file_name": {
"type": "string"
}
},
"required": [
"download_url",
"file_id"
],
"additionalProperties": false
}
},
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {
"status": {
"type": "string",
"const": "browser_analysis_started"
},
"analyzer_version": {
"type": "string"
},
"file": {
"type": "object",
"properties": {
"file_id": {
"type": "string"
},
"file_name": {
"type": [
"string",
"null"
]
},
"mime_type": {
"type": [
"string",
"null"
]
}
},
"required": [
"file_id",
"file_name",
"mime_type"
],
"additionalProperties": false
},
"analysis_depth": {
"type": "string"
},
"privacy": {
"type": "string"
}
},
"required": [
"status",
"analyzer_version",
"file",
"analysis_depth",
"privacy"
],
"additionalProperties": false
}🟢deepbom_capabilities
Use before analysis when format support, privacy boundaries, or the difference between the ChatGPT and local DEEPBOM paths is unclear. Returns capability metadata only and does not inspect a file.
输入模式
{
"type": "object",
"properties": {},
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {
"schema": {
"type": "string"
},
"version": {
"type": "string"
},
"supported_formats": {
"type": "array",
"items": {
"type": "string"
}
},
"chatgpt_path": {
"type": "object"
},
"local_path": {
"type": "object"
},
"evidence_boundary": {
"type": "string"
}
},
"required": [
"schema",
"version",
"supported_formats",
"chatgpt_path",
"local_path",
"evidence_boundary"
],
"additionalProperties": false
}🟢deepbom_publish_analysis(result)
Receives one bounded browser-produced DEEPBOM result, validates its structure, and returns it to the conversation. This tool accepts analysis results, never model bytes.
输入模式
{
"type": "object",
"properties": {
"result": {
"type": "object",
"anyOf": [
{
"type": "object",
"properties": {
"schema": {
"type": "string",
"const": "deepbom.chatgpt_analysis_result.v1"
},
"analyzer_version": {
"type": "string"
},
"analysis_location": {
"type": "string",
"const": "chatgpt_browser_sandbox"
},
"artifact": {
"type": "object",
"properties": {
"filename": {
"type": "string"
},
"format": {
"type": "string"
},
"sha256": {
"type": "string"
},
"byte_length": {
"type": "integer",
"minimum": 0
},
"artifact_ir_sha256": {
"type": [
"string",
"null"
]
}
},
"required": [
"filename",
"format",
"sha256",
"byte_length"
],
"additionalProperties": false
},
"verdict": {
"type": "object",
"properties": {
"status": {
"type": "string"
},
"artifact_defect_count": {
"type": "integer",
"minimum": 0
},
"caution_count": {
"type": "integer",
"minimum": 0
},
"evidence_needed_count": {
"type": "integer",
"minimum": 0
}
},
"required": [
"status",
"artifact_defect_count",
"caution_count",
"evidence_needed_count"
],
"additionalProperties": false
},
"graph": {
"type": "object",
"properties": {
"operator_count": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"tensor_count": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"total_macs": {
"type": [
"number",
"null"
],
"minimum": 0
},
"mac_confidence": {
"type": "string"
}
},
"required": [
"operator_count",
"tensor_count",
"total_macs",
"mac_confidence"
],
"additionalProperties": false
},
"storage": {
"type": [
"object",
"null"
]
},
"quantization": {
"type": "object"
},
"model_summary": {
"type": "object"
},
"findings": {
"type": "object"
},
"evidence_boundary": {
"type": "string"
},
"transfer_boundary": {
"type": "string"
},
"reproduction": {
"type": [
"object",
"null"
]
}
},
"required": [
"schema",
"analyzer_version",
"analysis_location",
"artifact",
"verdict",
"graph",
"storage",
"quantization",
"model_summary",
"findings",
"evidence_boundary",
"transfer_boundary",
"reproduction"
],
"additionalProperties": false
},
{
"type": "object",
"properties": {
"schema": {
"type": "string",
"const": "deepbom.chatgpt_analysis_result.v2"
},
"analyzer_version": {
"type": "string"
},
"analysis_location": {
"type": "string",
"const": "chatgpt_browser_sandbox"
},
"artifact": {
"type": "object",
"properties": {
"filename": {
"type": "string"
},
"format": {
"type": "string"
},
"sha256": {
"type": "string"
},
"byte_length": {
"type": "integer",
"minimum": 0
},
"artifact_ir_sha256": {
"type": [
"string",
"null"
]
}
},
"required": [
"filename",
"format",
"sha256",
"byte_length"
],
"additionalProperties": false
},
"verdict": {
"type": "object",
"properties": {
"status": {
"type": "string"
},
"artifact_defect_count": {
"type": "integer",
"minimum": 0
},
"caution_count": {
"type": "integer",
"minimum": 0
},
"evidence_needed_count": {
"type": "integer",
"minimum": 0
}
},
"required": [
"status",
"artifact_defect_count",
"caution_count",
"evidence_needed_count"
],
"additionalProperties": false
},
"graph": {
"type": "object",
"properties": {
"operator_count": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"tensor_count": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"total_macs": {
"type": [
"number",
"null"
],
"minimum": 0
},
"mac_confidence": {
"type": "string"
}
},
"required": [
"operator_count",
"tensor_count",
"total_macs",
"mac_confidence"
],
"additionalProperties": false
},
"storage": {
"type": [
"object",
"null"
]
},
"quantization": {
"type": "object"
},
"model_summary": {
"type": "object"
},
"provenance_summary": {
"type": "object",
"properties": {
"schema": {
"type": "string",
"const": "deepbom.provenance_summary.v1"
},
"artifact_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"model_ir_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"provenance_ir_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"status": {
"type": "string",
"enum": [
"contradiction_observed",
"incomplete",
"no_contradiction_observed"
]
},
"node_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"relationship_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"observed_file_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"check_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"field_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"mapped_declared_field_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"unsupported_field_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"attested_relationship_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"check_counts": {
"type": "object",
"properties": {
"match": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"mismatch": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"not_assessed": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"unresolved": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"unsupported": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
}
},
"required": [
"match",
"mismatch",
"not_assessed",
"unresolved",
"unsupported"
],
"additionalProperties": false
},
"metadata_truth_verified": {
"type": "boolean",
"const": false
},
"publisher_authenticity": {
"type": "string",
"const": "not_verified"
}
},
"required": [
"schema",
"artifact_sha256",
"model_ir_sha256",
"provenance_ir_sha256",
"status",
"node_count",
"relationship_count",
"observed_file_count",
"check_count",
"field_count",
"mapped_declared_field_count",
"unsupported_field_count",
"attested_relationship_count",
"check_counts",
"metadata_truth_verified",
"publisher_authenticity"
],
"additionalProperties": false
},
"findings": {
"type": "object"
},
"evidence_boundary": {
"type": "string"
},
"transfer_boundary": {
"type": "string"
},
"reproduction": {
"type": [
"object",
"null"
]
}
},
"required": [
"schema",
"analyzer_version",
"analysis_location",
"artifact",
"verdict",
"graph",
"storage",
"quantization",
"model_summary",
"findings",
"evidence_boundary",
"transfer_boundary",
"reproduction"
],
"additionalProperties": false
}
]
}
},
"required": [
"result"
],
"additionalProperties": false
}输出模式
{
"type": "object",
"anyOf": [
{
"type": "object",
"properties": {
"schema": {
"type": "string",
"const": "deepbom.chatgpt_analysis_result.v1"
},
"analyzer_version": {
"type": "string"
},
"analysis_location": {
"type": "string",
"const": "chatgpt_browser_sandbox"
},
"artifact": {
"type": "object",
"properties": {
"filename": {
"type": "string"
},
"format": {
"type": "string"
},
"sha256": {
"type": "string"
},
"byte_length": {
"type": "integer",
"minimum": 0
},
"artifact_ir_sha256": {
"type": [
"string",
"null"
]
}
},
"required": [
"filename",
"format",
"sha256",
"byte_length"
],
"additionalProperties": false
},
"verdict": {
"type": "object",
"properties": {
"status": {
"type": "string"
},
"artifact_defect_count": {
"type": "integer",
"minimum": 0
},
"caution_count": {
"type": "integer",
"minimum": 0
},
"evidence_needed_count": {
"type": "integer",
"minimum": 0
}
},
"required": [
"status",
"artifact_defect_count",
"caution_count",
"evidence_needed_count"
],
"additionalProperties": false
},
"graph": {
"type": "object",
"properties": {
"operator_count": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"tensor_count": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"total_macs": {
"type": [
"number",
"null"
],
"minimum": 0
},
"mac_confidence": {
"type": "string"
}
},
"required": [
"operator_count",
"tensor_count",
"total_macs",
"mac_confidence"
],
"additionalProperties": false
},
"storage": {
"type": [
"object",
"null"
]
},
"quantization": {
"type": "object"
},
"model_summary": {
"type": "object"
},
"findings": {
"type": "object"
},
"evidence_boundary": {
"type": "string"
},
"transfer_boundary": {
"type": "string"
},
"reproduction": {
"type": [
"object",
"null"
]
}
},
"required": [
"schema",
"analyzer_version",
"analysis_location",
"artifact",
"verdict",
"graph",
"storage",
"quantization",
"model_summary",
"findings",
"evidence_boundary",
"transfer_boundary",
"reproduction"
],
"additionalProperties": false
},
{
"type": "object",
"properties": {
"schema": {
"type": "string",
"const": "deepbom.chatgpt_analysis_result.v2"
},
"analyzer_version": {
"type": "string"
},
"analysis_location": {
"type": "string",
"const": "chatgpt_browser_sandbox"
},
"artifact": {
"type": "object",
"properties": {
"filename": {
"type": "string"
},
"format": {
"type": "string"
},
"sha256": {
"type": "string"
},
"byte_length": {
"type": "integer",
"minimum": 0
},
"artifact_ir_sha256": {
"type": [
"string",
"null"
]
}
},
"required": [
"filename",
"format",
"sha256",
"byte_length"
],
"additionalProperties": false
},
"verdict": {
"type": "object",
"properties": {
"status": {
"type": "string"
},
"artifact_defect_count": {
"type": "integer",
"minimum": 0
},
"caution_count": {
"type": "integer",
"minimum": 0
},
"evidence_needed_count": {
"type": "integer",
"minimum": 0
}
},
"required": [
"status",
"artifact_defect_count",
"caution_count",
"evidence_needed_count"
],
"additionalProperties": false
},
"graph": {
"type": "object",
"properties": {
"operator_count": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"tensor_count": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"total_macs": {
"type": [
"number",
"null"
],
"minimum": 0
},
"mac_confidence": {
"type": "string"
}
},
"required": [
"operator_count",
"tensor_count",
"total_macs",
"mac_confidence"
],
"additionalProperties": false
},
"storage": {
"type": [
"object",
"null"
]
},
"quantization": {
"type": "object"
},
"model_summary": {
"type": "object"
},
"provenance_summary": {
"type": "object",
"properties": {
"schema": {
"type": "string",
"const": "deepbom.provenance_summary.v1"
},
"artifact_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"model_ir_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"provenance_ir_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"status": {
"type": "string",
"enum": [
"contradiction_observed",
"incomplete",
"no_contradiction_observed"
]
},
"node_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"relationship_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"observed_file_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"check_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"field_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"mapped_declared_field_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"unsupported_field_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"attested_relationship_count": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"check_counts": {
"type": "object",
"properties": {
"match": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"mismatch": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"not_assessed": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"unresolved": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
},
"unsupported": {
"type": "integer",
"minimum": 0,
"maximum": 1000000
}
},
"required": [
"match",
"mismatch",
"not_assessed",
"unresolved",
"unsupported"
],
"additionalProperties": false
},
"metadata_truth_verified": {
"type": "boolean",
"const": false
},
"publisher_authenticity": {
"type": "string",
"const": "not_verified"
}
},
"required": [
"schema",
"artifact_sha256",
"model_ir_sha256",
"provenance_ir_sha256",
"status",
"node_count",
"relationship_count",
"observed_file_count",
"check_count",
"field_count",
"mapped_declared_field_count",
"unsupported_field_count",
"attested_relationship_count",
"check_counts",
"metadata_truth_verified",
"publisher_authenticity"
],
"additionalProperties": false
},
"findings": {
"type": "object"
},
"evidence_boundary": {
"type": "string"
},
"transfer_boundary": {
"type": "string"
},
"reproduction": {
"type": [
"object",
"null"
]
}
},
"required": [
"schema",
"analyzer_version",
"analysis_location",
"artifact",
"verdict",
"graph",
"storage",
"quantization",
"model_summary",
"findings",
"evidence_boundary",
"transfer_boundary",
"reproduction"
],
"additionalProperties": false
}
]
}🟢deepbom_publish_error(error)
Receives one bounded browser-produced DEEPBOM error and returns a recovery action to the conversation. This tool accepts diagnostic text, never model bytes.
输入模式
{
"type": "object",
"properties": {
"error": {
"type": "object",
"properties": {
"schema": {
"type": "string",
"const": "deepbom.chatgpt_error.v1"
},
"analyzer_version": {
"type": "string"
},
"analysis_location": {
"type": "string",
"const": "chatgpt_browser_sandbox"
},
"code": {
"type": "string"
},
"message": {
"type": "string"
},
"suggested_action": {
"type": "string"
},
"file_name": {
"type": [
"string",
"null"
]
},
"transfer_boundary": {
"type": "string",
"const": "model_bytes_not_sent_to_deepbom_service"
}
},
"required": [
"schema",
"analyzer_version",
"analysis_location",
"code",
"message",
"suggested_action",
"file_name",
"transfer_boundary"
],
"additionalProperties": false
}
},
"required": [
"error"
],
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {
"schema": {
"type": "string",
"const": "deepbom.chatgpt_error.v1"
},
"analyzer_version": {
"type": "string"
},
"analysis_location": {
"type": "string",
"const": "chatgpt_browser_sandbox"
},
"code": {
"type": "string"
},
"message": {
"type": "string"
},
"suggested_action": {
"type": "string"
},
"file_name": {
"type": [
"string",
"null"
]
},
"transfer_boundary": {
"type": "string",
"const": "model_bytes_not_sent_to_deepbom_service"
}
},
"required": [
"schema",
"analyzer_version",
"analysis_location",
"code",
"message",
"suggested_action",
"file_name",
"transfer_boundary"
],
"additionalProperties": false
}🟢deepbom_query_file(file, expected_sha256, analysis_depth, query)
Use for follow-up questions about an already identified attachment: inspect a layer by subject_ref, search or page through all operations, inspect findings, rank improvement investigations, inspect encoded activation fusion, compare source-pinned static delegate eligibility, or explicitly analyze weight distributions, channels, similarity, singular spectra, sparsity and quantization with section=weights and optional weight_view. Weight analysis reads payloads on demand with explicit decoder/budget coverage (128 MiB widget limit). Reuse the ChatGPT-authorized file and the exact SHA-256 from its previous DEEPBOM result. section=profiles lists available backend profiles; query.target selects only a TFLite CPU planning profile from capabilities. The browser computes the query from the shared Evidence IR and returns a bounded result; model bytes are not sent to DEEPBOM. Static candidate eligibility is not observed placement, speed, accuracy, or runtime fusion. Do not invent findings while browser_query_started is pending.
输入模式
{
"type": "object",
"properties": {
"file": {
"$ref": "#/$defs/OpenAIFile"
},
"expected_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"analysis_depth": {
"type": "string",
"enum": [
"structure",
"payload_integrity"
],
"default": "structure",
"description": "Use structure unless the user explicitly asks to inspect serialized tensor payload values. Payload integrity can require reading the complete attachment."
},
"query": {
"type": "object",
"additionalProperties": false,
"properties": {
"section": {
"type": "string",
"enum": [
"operators",
"operator",
"findings",
"placement",
"fusion",
"improvements",
"profiles",
"weights"
]
},
"subject_ref": {
"type": "string",
"minLength": 1,
"maxLength": 512,
"description": "Exact Model IR subject_ref from a previous result; preferred over an index."
},
"source_index": {
"type": "integer",
"minimum": 0,
"description": "Native serialized operator index, not display order. Ambiguous indices require subject_ref."
},
"search": {
"type": "string",
"maxLength": 160,
"description": "Literal case-insensitive filter over row names and references."
},
"offset": {
"type": "integer",
"minimum": 0,
"default": 0
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 40,
"default": 20
},
"profile_ids": {
"type": "array",
"items": {
"type": "string",
"minLength": 1,
"maxLength": 128
},
"uniqueItems": true,
"maxItems": 4,
"description": "Exact static backend profile IDs returned by section=profiles. Independent eligibility profiles, not runtime assignments."
},
"target": {
"type": "string",
"minLength": 1,
"maxLength": 128,
"description": "Optional TFLite CPU planning profile ID from capabilities. Not detected hardware or a GPU delegate selector."
},
"weight_view": {
"type": "string",
"enum": [
"distribution",
"channels",
"similarity",
"spectrum",
"sparsity",
"quantization"
],
"description": "Only for section=weights: explicitly opt into bounded payload analysis; use returned weight subject_ref to select a tensor. Default distribution."
}
},
"required": [
"section"
]
}
},
"required": [
"file",
"expected_sha256",
"query"
],
"$defs": {
"OpenAIFile": {
"type": "object",
"properties": {
"download_url": {
"type": "string"
},
"file_id": {
"type": "string"
},
"mime_type": {
"type": "string"
},
"file_name": {
"type": "string"
}
},
"required": [
"download_url",
"file_id"
],
"additionalProperties": false
}
},
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {
"status": {
"type": "string",
"const": "browser_query_started"
},
"analyzer_version": {
"type": "string"
},
"expected_sha256": {
"type": "string"
},
"analysis_depth": {
"type": "string",
"enum": [
"structure",
"payload_integrity"
],
"default": "structure",
"description": "Use structure unless the user explicitly asks to inspect serialized tensor payload values. Payload integrity can require reading the complete attachment."
},
"query": {
"type": "object",
"additionalProperties": false,
"properties": {
"section": {
"type": "string",
"enum": [
"operators",
"operator",
"findings",
"placement",
"fusion",
"improvements",
"profiles",
"weights"
]
},
"subject_ref": {
"type": "string",
"minLength": 1,
"maxLength": 512,
"description": "Exact Model IR subject_ref from a previous result; preferred over an index."
},
"source_index": {
"type": "integer",
"minimum": 0,
"description": "Native serialized operator index, not display order. Ambiguous indices require subject_ref."
},
"search": {
"type": "string",
"maxLength": 160,
"description": "Literal case-insensitive filter over row names and references."
},
"offset": {
"type": "integer",
"minimum": 0,
"default": 0
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 40,
"default": 20
},
"profile_ids": {
"type": "array",
"items": {
"type": "string",
"minLength": 1,
"maxLength": 128
},
"uniqueItems": true,
"maxItems": 4,
"description": "Exact static backend profile IDs returned by section=profiles. Independent eligibility profiles, not runtime assignments."
},
"target": {
"type": "string",
"minLength": 1,
"maxLength": 128,
"description": "Optional TFLite CPU planning profile ID from capabilities. Not detected hardware or a GPU delegate selector."
},
"weight_view": {
"type": "string",
"enum": [
"distribution",
"channels",
"similarity",
"spectrum",
"sparsity",
"quantization"
],
"description": "Only for section=weights: explicitly opt into bounded payload analysis; use returned weight subject_ref to select a tensor. Default distribution."
}
},
"required": [
"section"
]
}
},
"required": [
"status",
"analyzer_version",
"expected_sha256",
"analysis_depth",
"query"
],
"additionalProperties": false
}🟢deepbom_publish_query(result)
Component-only: validates and returns a bounded Evidence IR query projection for the same attachment. No model bytes, state persistence, or runtime execution. Result digests establish consistency, not independent server attestation.
输入模式
{
"type": "object",
"properties": {
"result": {
"type": "object",
"additionalProperties": false,
"properties": {
"schema": {
"const": "deepbom.evidence_query_result.v1",
"type": "string"
},
"analyzer_version": {
"type": "string"
},
"analysis_location": {
"const": "chatgpt_browser_sandbox",
"type": "string"
},
"artifact": {
"type": "object",
"additionalProperties": false,
"properties": {
"filename": {
"type": "string"
},
"format": {
"type": "string"
},
"sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
}
},
"required": [
"filename",
"format",
"sha256"
]
},
"model_ir_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"query": {
"type": "object",
"additionalProperties": false,
"properties": {
"section": {
"type": "string",
"enum": [
"operators",
"operator",
"findings",
"placement",
"fusion",
"improvements",
"profiles",
"weights"
]
},
"subject_ref": {
"type": "string",
"minLength": 1,
"maxLength": 512,
"description": "Exact Model IR subject_ref from a previous result; preferred over an index."
},
"source_index": {
"type": "integer",
"minimum": 0,
"description": "Native serialized operator index, not display order. Ambiguous indices require subject_ref."
},
"search": {
"type": "string",
"maxLength": 160,
"description": "Literal case-insensitive filter over row names and references."
},
"offset": {
"type": "integer",
"minimum": 0,
"default": 0
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 40,
"default": 20
},
"profile_ids": {
"type": "array",
"items": {
"type": "string",
"minLength": 1,
"maxLength": 128
},
"uniqueItems": true,
"maxItems": 4,
"description": "Exact static backend profile IDs returned by section=profiles. Independent eligibility profiles, not runtime assignments."
},
"target": {
"type": "string",
"minLength": 1,
"maxLength": 128,
"description": "Optional TFLite CPU planning profile ID from capabilities. Not detected hardware or a GPU delegate selector."
},
"weight_view": {
"type": "string",
"enum": [
"distribution",
"channels",
"similarity",
"spectrum",
"sparsity",
"quantization"
],
"description": "Only for section=weights: explicitly opt into bounded payload analysis; use returned weight subject_ref to select a tensor. Default distribution."
}
},
"required": [
"section"
]
},
"query_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"status": {
"type": "string",
"enum": [
"complete",
"partial",
"not_assessable",
"no_matches"
]
},
"coverage": {
"type": "object",
"additionalProperties": false,
"properties": {
"total_rows": {
"type": "integer",
"minimum": 0
},
"matched_rows": {
"type": "integer",
"minimum": 0
},
"returned_rows": {
"type": "integer",
"minimum": 0,
"maximum": 40
},
"offset": {
"type": "integer",
"minimum": 0
},
"next_offset": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"detail_truncated": {
"type": "boolean"
}
},
"required": [
"total_rows",
"matched_rows",
"returned_rows",
"offset",
"next_offset",
"detail_truncated"
]
},
"context": {
"type": "object"
},
"rows": {
"type": "array",
"maxItems": 40,
"items": {
"type": "object",
"additionalProperties": false,
"properties": {
"subject_ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"details": {
"type": "object"
},
"detail_truncations": {
"type": "array",
"items": {
"type": "string"
}
}
},
"required": [
"subject_ref",
"kind",
"title",
"details",
"detail_truncations"
]
}
},
"evidence_boundary": {
"type": "string"
},
"transfer_boundary": {
"type": "string",
"const": "model_bytes_not_sent_to_deepbom_service"
},
"result_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
}
},
"required": [
"schema",
"analyzer_version",
"analysis_location",
"artifact",
"model_ir_sha256",
"query",
"query_sha256",
"status",
"coverage",
"context",
"rows",
"evidence_boundary",
"transfer_boundary",
"result_sha256"
]
}
},
"required": [
"result"
],
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {
"schema": {
"const": "deepbom.evidence_query_result.v1",
"type": "string"
},
"analyzer_version": {
"type": "string"
},
"analysis_location": {
"const": "chatgpt_browser_sandbox",
"type": "string"
},
"artifact": {
"type": "object",
"additionalProperties": false,
"properties": {
"filename": {
"type": "string"
},
"format": {
"type": "string"
},
"sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
}
},
"required": [
"filename",
"format",
"sha256"
]
},
"model_ir_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"query": {
"type": "object",
"additionalProperties": false,
"properties": {
"section": {
"type": "string",
"enum": [
"operators",
"operator",
"findings",
"placement",
"fusion",
"improvements",
"profiles",
"weights"
]
},
"subject_ref": {
"type": "string",
"minLength": 1,
"maxLength": 512,
"description": "Exact Model IR subject_ref from a previous result; preferred over an index."
},
"source_index": {
"type": "integer",
"minimum": 0,
"description": "Native serialized operator index, not display order. Ambiguous indices require subject_ref."
},
"search": {
"type": "string",
"maxLength": 160,
"description": "Literal case-insensitive filter over row names and references."
},
"offset": {
"type": "integer",
"minimum": 0,
"default": 0
},
"limit": {
"type": "integer",
"minimum": 1,
"maximum": 40,
"default": 20
},
"profile_ids": {
"type": "array",
"items": {
"type": "string",
"minLength": 1,
"maxLength": 128
},
"uniqueItems": true,
"maxItems": 4,
"description": "Exact static backend profile IDs returned by section=profiles. Independent eligibility profiles, not runtime assignments."
},
"target": {
"type": "string",
"minLength": 1,
"maxLength": 128,
"description": "Optional TFLite CPU planning profile ID from capabilities. Not detected hardware or a GPU delegate selector."
},
"weight_view": {
"type": "string",
"enum": [
"distribution",
"channels",
"similarity",
"spectrum",
"sparsity",
"quantization"
],
"description": "Only for section=weights: explicitly opt into bounded payload analysis; use returned weight subject_ref to select a tensor. Default distribution."
}
},
"required": [
"section"
]
},
"query_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
},
"status": {
"type": "string",
"enum": [
"complete",
"partial",
"not_assessable",
"no_matches"
]
},
"coverage": {
"type": "object",
"additionalProperties": false,
"properties": {
"total_rows": {
"type": "integer",
"minimum": 0
},
"matched_rows": {
"type": "integer",
"minimum": 0
},
"returned_rows": {
"type": "integer",
"minimum": 0,
"maximum": 40
},
"offset": {
"type": "integer",
"minimum": 0
},
"next_offset": {
"type": [
"integer",
"null"
],
"minimum": 0
},
"detail_truncated": {
"type": "boolean"
}
},
"required": [
"total_rows",
"matched_rows",
"returned_rows",
"offset",
"next_offset",
"detail_truncated"
]
},
"context": {
"type": "object"
},
"rows": {
"type": "array",
"maxItems": 40,
"items": {
"type": "object",
"additionalProperties": false,
"properties": {
"subject_ref": {
"type": "string"
},
"kind": {
"type": "string"
},
"title": {
"type": "string"
},
"details": {
"type": "object"
},
"detail_truncations": {
"type": "array",
"items": {
"type": "string"
}
}
},
"required": [
"subject_ref",
"kind",
"title",
"details",
"detail_truncations"
]
}
},
"evidence_boundary": {
"type": "string"
},
"transfer_boundary": {
"type": "string",
"const": "model_bytes_not_sent_to_deepbom_service"
},
"result_sha256": {
"type": "string",
"pattern": "^[a-f0-9]{64}$"
}
},
"required": [
"schema",
"analyzer_version",
"analysis_location",
"artifact",
"model_ir_sha256",
"query",
"query_sha256",
"status",
"coverage",
"context",
"rows",
"evidence_boundary",
"transfer_boundary",
"result_sha256"
],
"additionalProperties": false
}🟢deepbom_workflow_guide(workflow, format)
Use to discover how to answer a model-analysis question across ChatGPT, the interactive deepbom.org workspace and the local CLI. Returns real query sections, visualization/export actions, explicit web continuation links, required inputs, CLI argument templates and evidence limits for weights, pruning investigation, comparison, BOM verification, metadata/OMOP, activation evidence, runtime, batch/CI and integrations. No attachment is required. This is guidance, not a completed analysis or local command execution. Prefer answering from existing evidence; inspect supported follow-ups with deepbom_query_file before offering web continuation for richer interaction. Website links transfer only a workflow ID, never model data.
输入模式
{
"type": "object",
"properties": {
"workflow": {
"type": "string",
"enum": [
"all",
"audit",
"layers",
"findings",
"quantization",
"placement",
"fusion",
"visualization",
"weights",
"pruning",
"compare",
"redesign",
"exports",
"verify",
"provenance",
"activation",
"runtime",
"batch",
"integrations"
],
"default": "all"
},
"format": {
"type": "string",
"enum": [
"tflite",
"onnx",
"gguf",
"safetensors",
"coreml",
"executorch"
],
"description": "Optional known artifact format; filters applicable workflows, not per-file analysis coverage."
}
},
"additionalProperties": false
}输出模式
{
"type": "object",
"properties": {
"schema": {
"type": "string"
},
"catalog_version": {
"type": "string"
},
"status": {
"type": "string"
},
"format": {
"type": [
"string",
"null"
]
},
"workflows": {
"type": "array",
"items": {
"type": "object"
}
},
"boundary": {
"type": "string"
}
},
"required": [
"schema",
"catalog_version",
"status",
"format",
"workflows",
"boundary"
],
"additionalProperties": false
}社区
证据
最近观测
已验证未记录版本7 个工具
已验证未记录版本4 个工具
已验证未记录版本4 个工具