{"schema_version":1,"workflow_uuid":"e0a9738b-37db-11f1-9bc6-00163e2b0d79","workflow_title":"NVIDIA Triton Inference Server — Multi-Framework Model Serving at Scale","page_url":"https://tokrepo.com/en/workflows/nvidia-triton-inference-server-multi-framework-model-e0a9738b","raw_url":"https://tokrepo.com/raw/nvidia-triton-inference-server-multi-framework-model-e0a9738b","metadata_url":"https://tokrepo.com/metadata/nvidia-triton-inference-server-multi-framework-model-e0a9738b.json","install_plan_url":"https://api.tokrepo.com/api/v1/tokenboard/workflows/install-plan?uuid=e0a9738b-37db-11f1-9bc6-00163e2b0d79&target=codex","recommended_install":{"schema_version":1,"status":"stage","policy":"stage_only","install_mode":"stage_only","target":"codex","command":"npx -y tokrepo@latest install e0a9738b-37db-11f1-9bc6-00163e2b0d79 --target codex","dry_run_command":"npx -y tokrepo@latest install e0a9738b-37db-11f1-9bc6-00163e2b0d79 --target codex --dry-run --json","runtime_commands":{"codex":"npx -y tokrepo@latest install e0a9738b-37db-11f1-9bc6-00163e2b0d79 --target codex","gemini":"npx -y tokrepo@latest install e0a9738b-37db-11f1-9bc6-00163e2b0d79 --target gemini","project":"npx -y tokrepo@latest install e0a9738b-37db-11f1-9bc6-00163e2b0d79"},"prompt":"Install and make this TokRepo asset usable: \"NVIDIA Triton Inference Server — Multi-Framework Model Serving at Scale\".\n\nPage: https://tokrepo.com/en/workflows/nvidia-triton-inference-server-multi-framework-model-e0a9738b\nUUID: e0a9738b-37db-11f1-9bc6-00163e2b0d79\nTarget adapter: codex\nPolicy/mode: stage_only / stage_only\n\nRead these machine-facing entry points first:\n- metadata: https://tokrepo.com/metadata/nvidia-triton-inference-server-multi-framework-model-e0a9738b.json\n- install contract: https://tokrepo.com/install-contract/nvidia-triton-inference-server-multi-framework-model-e0a9738b.json\n- install plan: https://api.tokrepo.com/api/v1/tokenboard/workflows/install-plan?uuid=e0a9738b-37db-11f1-9bc6-00163e2b0d79&target=codex\n- raw content: https://tokrepo.com/raw/nvidia-triton-inference-server-multi-framework-model-e0a9738b\n\nCommands:\n- choose the current runtime; do not use unsupported CLI targets.\n- Codex: npx -y tokrepo@latest install e0a9738b-37db-11f1-9bc6-00163e2b0d79 --target codex\n- Gemini CLI: npx -y tokrepo@latest install e0a9738b-37db-11f1-9bc6-00163e2b0d79 --target gemini\n- Claude Code / project agent: run npx -y tokrepo@latest install e0a9738b-37db-11f1-9bc6-00163e2b0d79 from the project root; add `--dry-run --json` first to inspect the plan.\n- add `--dry-run --json` to the chosen command first; after the plan allows it, run the chosen command.\n\nExecution rules:\n1. Run the install command to stage the asset safely.\n2. Read the staged README, install-plan, and entrypoint files.\n3. Activate scripts, MCP config, or global config only after user confirmation.\n4. Verify usability with the install-plan post_verify checks and the asset README.\n\nReport back with changed files, verification result, and how to use the asset next.","next_steps":["Run the install command to stage the asset safely.","Read the staged README, install-plan, and entrypoint files.","Activate scripts, MCP config, or global config only after user confirmation.","Verify usability with the install-plan post_verify checks and the asset README."],"success_check":["The asset is safely staged.","The agent can give clear activation steps from the staged content."]},"agent_metadata":{"asset_kind":"skill","target_tools":["claude_code","codex","gemini_cli"],"install_mode":"single","entrypoint":"step-1.md","risk_profile":{"executes_code":false,"modifies_global_config":false,"requires_secrets":[],"uses_absolute_paths":false,"network_access":true},"dependencies":{"npm":[],"pip":[],"brew":[],"system":[]},"content_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","verification":{"commands":[],"expected_files":[""]}},"agent_fit":{"target":"codex","score":29,"status":"stage_only","policy":"stage_only","why":["target_tools includes codex","asset_kind skill","install_mode stage_only","policy stage_only","install_mode is stage_only","risk_profile.network_access is true","trust community"],"asset_kind":"skill","install_mode":"stage_only"},"trust":{"author_trust_level":"community","verified_publisher":false,"asset_signed_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","signature_status":"hash_only","install_count":0,"report_count":0,"dangerous_capability_badges":["network_access"],"review_status":"unreviewed","signals":["asset has usage views","content hash available"]},"provenance":{"owner_uuid":"70b84b45-463b-11f1-9bc6-00163e2b0d79","owner_name":"NVIDIA","source_url":"https://tokrepo.com/en/workflows/nvidia-triton-inference-server-multi-framework-model-e0a9738b","content_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","visibility":1,"created_at":"2026-04-14 16:28:13","updated_at":"2026-06-13 18:26:41"},"target_adapter":{"target":"codex","adapter":"skill-directory","root":"~/.codex/skills","entrypoint":"SKILL.md","manifest_path":"~/.codex/tokrepo/install-manifest.json","staging_root":"~/.codex/tokrepo/staged/e0a9738b-37db-11f1-9bc6-00163e2b0d79","install_modes":["single","bundle","split","stage_only"],"activates_files":true},"install_plan":{"schema_version":2,"target":"codex","asset_uuid":"e0a9738b-37db-11f1-9bc6-00163e2b0d79","asset_title":"NVIDIA Triton Inference Server — Multi-Framework Model Serving at Scale","source_url":"https://tokrepo.com/en/workflows/nvidia-triton-inference-server-multi-framework-model-e0a9738b","install_mode":"stage_only","entrypoint":"step-1.md","preconditions":[{"type":"target_supported","status":"pass","message":"codex install target is supported"},{"type":"install_root","status":"pass","message":"~/.codex/skills for activated assets; ~/.codex/tokrepo/staged/e0a9738b-37db-11f1-9bc6-00163e2b0d79 for staged assets"},{"type":"target_tool_metadata","status":"pass","message":"metadata allows codex"},{"type":"content_hash","status":"pass","message":"asset metadata includes content_hash"},{"type":"trust_policy","status":"warn","message":"low trust publisher plus dangerous capability requires staging"},{"type":"policy_decision","status":"warn","message":"stage_only for e0a9738b-37db-11f1-9bc6-00163e2b0d79 (stage_only)"}],"actions":[{"type":"stage_file","path":"~/.codex/tokrepo/staged/e0a9738b-37db-11f1-9bc6-00163e2b0d79/step-1.md","source_name":"step-1.md","sha256":"e3b1e30aa6a34790016317b18326b261d0f2cd10b81953eab2c20dbd97f802fa","bytes":5323,"risk":{"executes_code":false,"modifies_global_config":false,"requires_secrets":[],"uses_absolute_paths":false,"network_access":true},"if_exists":"overwrite"}],"policy_decision":{"decision":"stage_only","requires_confirmation":false,"reasons":["install_mode is stage_only","risk_profile.network_access is true","low trust publisher plus dangerous capability requires staging"]},"requires_confirmation":false,"rollback":[{"type":"remove_file","path":"~/.codex/tokrepo/staged/e0a9738b-37db-11f1-9bc6-00163e2b0d79/step-1.md"}],"post_verify":[{"type":"file_sha256","path":"~/.codex/tokrepo/staged/e0a9738b-37db-11f1-9bc6-00163e2b0d79/step-1.md","sha256":"e3b1e30aa6a34790016317b18326b261d0f2cd10b81953eab2c20dbd97f802fa"}],"metadata":{"asset_kind":"skill","target_tools":["claude_code","codex","gemini_cli"],"install_mode":"stage_only","entrypoint":"step-1.md","risk_profile":{"executes_code":false,"modifies_global_config":false,"requires_secrets":[],"uses_absolute_paths":false,"network_access":true},"dependencies":{"npm":[],"pip":[],"brew":[],"system":[]},"content_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","verification":{"commands":[],"expected_files":[""]}},"agent_fit":{"target":"codex","score":29,"status":"stage_only","policy":"stage_only","why":["target_tools includes codex","asset_kind skill","install_mode stage_only","policy stage_only","install_mode is stage_only","risk_profile.network_access is true","trust community"],"asset_kind":"skill","install_mode":"stage_only"},"trust":{"author_trust_level":"community","verified_publisher":false,"asset_signed_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","signature_status":"hash_only","install_count":0,"report_count":0,"dangerous_capability_badges":["network_access"],"review_status":"unreviewed","signals":["asset has usage views","content hash available"]},"provenance":{"owner_uuid":"70b84b45-463b-11f1-9bc6-00163e2b0d79","owner_name":"NVIDIA","source_url":"https://tokrepo.com/en/workflows/nvidia-triton-inference-server-multi-framework-model-e0a9738b","content_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","visibility":1,"created_at":"2026-04-14 16:28:13","updated_at":"2026-06-13 18:26:41"},"target_adapter":{"target":"codex","adapter":"skill-directory","root":"~/.codex/skills","entrypoint":"SKILL.md","manifest_path":"~/.codex/tokrepo/install-manifest.json","staging_root":"~/.codex/tokrepo/staged/e0a9738b-37db-11f1-9bc6-00163e2b0d79","install_modes":["single","bundle","split","stage_only"],"activates_files":true},"evidence_bundle":{"acceptance_gate":{"recommended_action":"stage_or_request_confirmation","rule":"Agents should only activate an asset after evidence_bundle.integrity, policy_compatibility, rollback, and post_verify have been inspected.","status":"caution"},"asset_title":"NVIDIA Triton Inference Server — Multi-Framework Model Serving at Scale","asset_uuid":"e0a9738b-37db-11f1-9bc6-00163e2b0d79","eval_evidence":["https://tokrepo.com/evals/install-safety.json","https://tokrepo.com/evals/trust-evidence-coverage.json","https://tokrepo.com/evals/handoff-quality.json"],"generated_at":"2026-06-13T10:26:42Z","integrity":{"content_hash":"c6762de69559722a1003ca6f919baa5884ccad390b686520546ebccdeca438a8","declared_content_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","file_count":1,"hash_algorithm":"sha256","install_plan_hash":"eaa917d8881f23d599458e5fb4fa667d493c51de449d28597c4ef139e091836b"},"policy_compatibility":{"permission_envelope":{"destructive":false,"executes_code":false,"file_count":1,"filesystem_write":["~/.codex/tokrepo/staged"],"global_config_write":false,"network":true,"requires_secrets":[],"uses_absolute_paths":false},"policy_decision":{"decision":"stage_only","requires_confirmation":false,"reasons":["install_mode is stage_only","risk_profile.network_access is true","low trust publisher plus dangerous capability requires staging"]},"requires_confirmation":false,"target":"codex","trust_score_v2":{"recommended_action":"stage_or_request_confirmation","status":"caution","trust_score":60}},"provenance":{"asset_kind":"skill","asset_title":"NVIDIA Triton Inference Server — Multi-Framework Model Serving at Scale","asset_uuid":"e0a9738b-37db-11f1-9bc6-00163e2b0d79","computed_bundle_hash":"c6762de69559722a1003ca6f919baa5884ccad390b686520546ebccdeca438a8","content_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","created_at":"2026-04-14 16:28:13","generated_at":"2026-06-13T10:26:42Z","install_plan_hash":"eaa917d8881f23d599458e5fb4fa667d493c51de449d28597c4ef139e091836b","owner_name":"NVIDIA","owner_uuid":"70b84b45-463b-11f1-9bc6-00163e2b0d79","parent_uuid":"","schema_version":2,"source":"tokrepo_asset","source_url":"https://tokrepo.com/en/workflows/nvidia-triton-inference-server-multi-framework-model-e0a9738b","updated_at":"2026-06-13 18:26:41","visibility":1},"sbom":{"asset_kind":"skill","asset_title":"NVIDIA Triton Inference Server — Multi-Framework Model Serving at Scale","asset_uuid":"e0a9738b-37db-11f1-9bc6-00163e2b0d79","capability_flags":{"destructive":false,"executes_code":false,"modifies_global_config":false,"network_access":true,"requires_secrets":[]},"content_hash":"c6762de69559722a1003ca6f919baa5884ccad390b686520546ebccdeca438a8","dependencies":{"brew":[],"mcp":[],"npm":[],"pip":[],"system":[]},"files":[{"bytes":5323,"path":"~/.codex/tokrepo/staged/e0a9738b-37db-11f1-9bc6-00163e2b0d79/step-1.md","role":"supporting_file","sha256":"e3b1e30aa6a34790016317b18326b261d0f2cd10b81953eab2c20dbd97f802fa","source_name":"step-1.md"}],"format":"SBOM-lite","install_mode":"stage_only","schema_version":1,"target":"codex"},"schema":"https://tokrepo.com/schemas/agent-evidence-bundle.schema.json","schema_version":1,"schemas":{"asset_verification":"https://tokrepo.com/schemas/asset-verification.schema.json","evidence_bundle":"https://tokrepo.com/schemas/agent-evidence-bundle.schema.json","install_plan":"https://tokrepo.com/schemas/install-plan.schema.json","provenance":"https://tokrepo.com/schemas/provenance.schema.json","sbom":"https://tokrepo.com/schemas/agent-evidence-bundle.schema.json#/properties/sbom"},"signature_evidence":{"content_hash":"c6762de69559722a1003ca6f919baa5884ccad390b686520546ebccdeca438a8","hash_algorithm":"sha256","install_plan_hash":"eaa917d8881f23d599458e5fb4fa667d493c51de449d28597c4ef139e091836b","schema_version":1,"signed_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","status":"hash_only","verification_notes":["hash_only evidence proves content integrity but not publisher identity unless an external signature verifies it"]},"source_url":"https://tokrepo.com/en/workflows/nvidia-triton-inference-server-multi-framework-model-e0a9738b","target":"codex"},"sbom":{"asset_kind":"skill","asset_title":"NVIDIA Triton Inference Server — Multi-Framework Model Serving at Scale","asset_uuid":"e0a9738b-37db-11f1-9bc6-00163e2b0d79","capability_flags":{"destructive":false,"executes_code":false,"modifies_global_config":false,"network_access":true,"requires_secrets":[]},"content_hash":"c6762de69559722a1003ca6f919baa5884ccad390b686520546ebccdeca438a8","dependencies":{"brew":[],"mcp":[],"npm":[],"pip":[],"system":[]},"files":[{"bytes":5323,"path":"~/.codex/tokrepo/staged/e0a9738b-37db-11f1-9bc6-00163e2b0d79/step-1.md","role":"supporting_file","sha256":"e3b1e30aa6a34790016317b18326b261d0f2cd10b81953eab2c20dbd97f802fa","source_name":"step-1.md"}],"format":"SBOM-lite","install_mode":"stage_only","schema_version":1,"target":"codex"},"signature_evidence":{"content_hash":"c6762de69559722a1003ca6f919baa5884ccad390b686520546ebccdeca438a8","hash_algorithm":"sha256","install_plan_hash":"eaa917d8881f23d599458e5fb4fa667d493c51de449d28597c4ef139e091836b","schema_version":1,"signed_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","status":"hash_only","verification_notes":["hash_only evidence proves content integrity but not publisher identity unless an external signature verifies it"]},"provenance_v2":{"asset_kind":"skill","asset_title":"NVIDIA Triton Inference Server — Multi-Framework Model Serving at Scale","asset_uuid":"e0a9738b-37db-11f1-9bc6-00163e2b0d79","computed_bundle_hash":"c6762de69559722a1003ca6f919baa5884ccad390b686520546ebccdeca438a8","content_hash":"e52c35715c70366131c491119f2a86738bf0c73df7f5718e3af358c5a74eeebf","created_at":"2026-04-14 16:28:13","generated_at":"2026-06-13T10:26:42Z","install_plan_hash":"eaa917d8881f23d599458e5fb4fa667d493c51de449d28597c4ef139e091836b","owner_name":"NVIDIA","owner_uuid":"70b84b45-463b-11f1-9bc6-00163e2b0d79","parent_uuid":"","schema_version":2,"source":"tokrepo_asset","source_url":"https://tokrepo.com/en/workflows/nvidia-triton-inference-server-multi-framework-model-e0a9738b","updated_at":"2026-06-13 18:26:41","visibility":1},"transitive_dependencies":null}}