[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7904CSGKCYZQQZCWMG46YM2":3,"skill-same:01M7904CSGKCYZQQZCWMG46YM2":71},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":46},"01M7904CSGKCYZQQZCWMG46YM2","serving-and-cli",null,"Operate Mellea 0.8.0.dev0's m CLI and OpenAI-compatible server safely, including serving, routing, response schemas, streaming, decomposition, migration, evaluation routing, and bounded adapter commands.","VectorSpaceLab\u002FAREX-Skill","VectorSpaceLab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F180637102?v=4","01M7538C5WPSJGWF7KHR4EFZ9F","01M7538C6R2GFP53NPXTCZNBRT",351,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmellea\u002Fsub-skills\u002Fserving-and-cli\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: serving-and-cli\ndescription: \"Operate Mellea 0.8.0.dev0's m CLI and OpenAI-compatible server\n  safely, including serving, routing, response schemas, streaming,\n  decomposition, migration, evaluation routing, and bounded adapter commands.\"\ndisable-model-invocation: true\nmetadata:\n  disco-role: operating\nlicense: Apache 2.0\n---\n\n# Serving and CLI\n\nUse this route for the `m` entry point, `m serve`,\n`POST \u002Fv1\u002Fchat\u002Fcompletions`, `GET \u002Fhealth`, OpenAI-compatible clients, model\nrouting, response formats, SSE streaming, `m decompose`, `m fix`, or the\nbounded CLI aspects of `m alora` and `m eval`. The operating contract is for\nMellea `0.8.0.dev0` and Python 3.11+.\n\n## Route the task\n\n1. Run the safe bundled checker before invoking a side-effectful command:\n   `python scripts\u002Fcheck_cli_surface.py --mode static`. To render installed\n   help without executing a callback, use `--mode help --target \u003Cname>`.\n2. Read [CLI reference](references\u002Fcli-reference.md) for exact command families,\n   current defaults, output files, prerequisites, and side effects.\n3. For `m serve`, request\u002Fresponse fields, structured output, routing, streaming,\n   multimodal messages, and error envelopes, read\n   [Serving API](references\u002Fserving-api.md).\n4. Before binding a socket or exposing a process, read\n   [Deployment and configuration](references\u002Fdeployment-and-config.md).\n5. For missing extras, schema failures, port conflicts, backend errors, partial\n   output, or command-safety recovery, read\n   [Troubleshooting](references\u002Ftroubleshooting.md).\n\n## Safety gates\n\n- Help and static inspection are safe. `m serve` imports user code and starts a\n  blocking server; `m decompose` and `m eval` call models and write files;\n  `m fix genslots` writes unless `--dry-run`; aLoRA commands may download a\n  model, consume substantial compute, overwrite local files, prompt, or upload\n  to Hugging Face Hub.\n- Require an explicit decision before a public bind, provider call, model\n  download, training run, remote upload, or non-dry-run rewrite. Never use a\n  side-effectful command as an installation probe.\n- Prefer `127.0.0.1` over the `m serve` wildcard default. The built-in service\n  has no authentication, TLS, rate limiting, or authorization layer.\n- Treat request `model` as untrusted routing metadata. Route only through a\n  fixed allowlist; never derive a provider URL, import, credential, or arbitrary\n  model loader directly from it.\n- Validate `json_schema` before generation. A response schema affects output\n  only when the served function accepts `format` and passes it into Mellea.\n- True incremental streaming needs an uncomputed thunk and a streaming-capable\n  backend. A computed result becomes one SSE content chunk.\n\n## Minimal workflows\n\n### Plan a serving request\n\n1. Choose a local host and an unused port; prefer `127.0.0.1:8080`.\n2. Confirm the app's `serve` signature and backend prerequisites without\n   importing untrusted, side-effectful application code.\n3. Draft one minimal request with `model` and a user message.\n4. Add exactly one advanced feature at a time: `stream`, `response_format`,\n   tools, multimodal content, or model routing.\n5. Check `\u002Fhealth`, then send the bounded request and inspect the response or\n   SSE `[DONE]` marker.\n\n### Plan a CLI migration or pipeline\n\n1. Render the exact nested help surface and check paths\u002Fcredentials first.\n2. For `fix`, run `--dry-run` and review every match before writing.\n3. For `decompose` or `eval`, bound model\u002Ftoken\u002Ftimeout settings and use a new\n   output path; these commands make provider calls and files.\n4. Inspect generated Python, validators, and result counts before executing or\n   treating a run as successful.\n5. Record which operations were deferred because they need a live backend,\n   training resources, or remote write approval.\n\n### Route a difficult request\n\n- Structured-output endpoint with routing and streaming: read all of\n  [Serving API](references\u002Fserving-api.md), use a fixed route allowlist, validate\n  the JSON schema, and ensure the streaming branch returns an uncomputed thunk.\n- Missing server extra: run help\u002Fstatic checks, install `server`, and repeat\n  help; do not start Uvicorn as a dependency test.\n- Decompose\u002Ffix misuse: read the failure table in\n  [Troubleshooting](references\u002Ftroubleshooting.md), preserve generated or\n  rewritten artifacts, and recover with a narrower, reviewable operation.\n\n## Ownership boundaries\n\nRoute provider selection, model identifiers, credentials, backend extras, and\nprovider failures to `..\u002Fbackends-and-models\u002FSKILL.md`. Route evaluation design,\njudge validity, and sampling semantics to\n`..\u002Fsampling-and-evaluation\u002FSKILL.md`. Route tool execution, tool-call loops,\nand MCP to `..\u002Ftools-and-agents\u002FSKILL.md` when that route is available; this\nskill covers only the HTTP envelope and CLI boundary for tools.\n","# Serving and CLI\n\nUse this route for the `m` entry point, `m serve`,\n`POST \u002Fv1\u002Fchat\u002Fcompletions`, `GET \u002Fhealth`, OpenAI-compatible clients, model\nrouting, response formats, SSE streaming, `m decompose`, `m fix`, or the\nbounded CLI aspects of `m alora` and `m eval`. The operating contract is for\nMellea `0.8.0.dev0` and Python 3.11+.\n\n## Route the task\n\n1. Run the safe bundled checker before invoking a side-effectful command:\n   `python scripts\u002Fcheck_cli_surface.py --mode static`. To render installed\n   help without executing a callback, use `--mode help --target \u003Cname>`.\n2. Read [CLI reference](references\u002Fcli-reference.md) for exact command families,\n   current defaults, output files, prerequisites, and side effects.\n3. For `m serve`, request\u002Fresponse fields, structured output, routing, streaming,\n   multimodal messages, and error envelopes, read\n   [Serving API](references\u002Fserving-api.md).\n4. Before binding a socket or exposing a process, read\n   [Deployment and configuration](references\u002Fdeployment-and-config.md).\n5. For missing extras, schema failures, port conflicts, backend errors, partial\n   output, or command-safety recovery, read\n   [Troubleshooting](references\u002Ftroubleshooting.md).\n\n## Safety gates\n\n- Help and static inspection are safe. `m serve` imports user code and starts a\n  blocking server; `m decompose` and `m eval` call models and write files;\n  `m fix genslots` writes unless `--dry-run`; aLoRA commands may download a\n  model, consume substantial compute, overwrite local files, prompt, or upload\n  to Hugging Face Hub.\n- Require an explicit decision before a public bind, provider call, model\n  download, training run, remote upload, or non-dry-run rewrite. Never use a\n  side-effectful command as an installation probe.\n- Prefer `127.0.0.1` over the `m serve` wildcard default. The built-in service\n  has no authentication, TLS, rate limiting, or authorization layer.\n- Treat request `model` as untrusted routing metadata. Route only through a\n  fixed allowlist; never derive a provider URL, import, credential, or arbitrary\n  model loader directly from it.\n- Validate `json_schema` before generation. A response schema affects output\n  only when the served function accepts `format` and passes it into Mellea.\n- True incremental streaming needs an uncomputed thunk and a streaming-capable\n  backend. A computed result becomes one SSE content chunk.\n\n## Minimal workflows\n\n### Plan a serving request\n\n1. Choose a local host and an unused port; prefer `127.0.0.1:8080`.\n2. Confirm the app's `serve` signature and backend prerequisites without\n   importing untrusted, side-effectful application code.\n3. Draft one minimal request with `model` and a user message.\n4. Add exactly one advanced feature at a time: `stream`, `response_format`,\n   tools, multimodal content, or model routing.\n5. Check `\u002Fhealth`, then send the bounded request and inspect the response or\n   SSE `[DONE]` marker.\n\n### Plan a CLI migration or pipeline\n\n1. Render the exact nested help surface and check paths\u002Fcredentials first.\n2. For `fix`, run `--dry-run` and review every match before writing.\n3. For `decompose` or `eval`, bound model\u002Ftoken\u002Ftimeout settings and use a new\n   output path; these commands make provider calls and files.\n4. Inspect generated Python, validators, and result counts before executing or\n   treating a run as successful.\n5. Record which operations were deferred because they need a live backend,\n   training resources, or remote write approval.\n\n### Route a difficult request\n\n- Structured-output endpoint with routing and streaming: read all of\n  [Serving API](references\u002Fserving-api.md), use a fixed route allowlist, validate\n  the JSON schema, and ensure the streaming branch returns an uncomputed thunk.\n- Missing server extra: run help\u002Fstatic checks, install `server`, and repeat\n  help; do not start Uvicorn as a dependency test.\n- Decompose\u002Ffix misuse: read the failure table in\n  [Troubleshooting](references\u002Ftroubleshooting.md), preserve generated or\n  rewritten artifacts, and recover with a narrower, reviewable operation.\n\n## Ownership boundaries\n\nRoute provider selection, model identifiers, credentials, backend extras, and\nprovider failures to `..\u002Fbackends-and-models\u002FSKILL.md`. Route evaluation design,\njudge validity, and sampling semantics to\n`..\u002Fsampling-and-evaluation\u002FSKILL.md`. Route tool execution, tool-call loops,\nand MCP to `..\u002Ftools-and-agents\u002FSKILL.md` when that route is available; this\nskill covers only the HTTP envelope and CLI boundary for tools.\n",{"name":5,"license":23,"metadata":24,"description":7,"disable-model-invocation":26},"Apache 2.0",{"disco-role":25},"operating",true,"main","ac3fe1afa80fb9a09775ecfb2b6cc3ba850a2db6","skills\u002Frepositories\u002Frepo-skills\u002Fmellea\u002Fsub-skills\u002Fserving-and-cli",27,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab","2026-09-03 10:25:21",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":26},[36,39,42],{"tool":37,"command":38},"npx","npx skills add VectorSpaceLab\u002FAREX-Skill --skill serving-and-cli",{"tool":40,"command":41},"bunx","bunx skills add VectorSpaceLab\u002FAREX-Skill --skill serving-and-cli",{"tool":43,"command":44},"pnpm","pnpm dlx skills add VectorSpaceLab\u002FAREX-Skill --skill serving-and-cli","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmellea\u002Fsub-skills\u002Fserving-and-cli\u002FSKILL.md",[47,53,57,61,65],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"references\u002Fcli-reference.md","REFERENCE","参考文档",10672,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmellea\u002Fsub-skills\u002Fserving-and-cli\u002Freferences\u002Fcli-reference.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fdeployment-and-config.md",7785,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmellea\u002Fsub-skills\u002Fserving-and-cli\u002Freferences\u002Fdeployment-and-config.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fserving-api.md",11927,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmellea\u002Fsub-skills\u002Fserving-and-cli\u002Freferences\u002Fserving-api.md",{"path":62,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Ftroubleshooting.md",13424,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmellea\u002Fsub-skills\u002Fserving-and-cli\u002Freferences\u002Ftroubleshooting.md",{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"scripts\u002Fcheck_cli_surface.py","SCRIPT","脚本",9816,"https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Frepositories\u002Frepo-skills\u002Fmellea\u002Fsub-skills\u002Fserving-and-cli\u002Fscripts\u002Fcheck_cli_surface.py",[72,79,86,93,100,107],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":18},"01M799NYSH1R3RVAHSM2D88Y84","passnet-triton-opt","Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block\u002Fgrid\u002Fwarp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-triton-opt\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":18},"01M799NS5APEC0N1FZ7XZVW6BN","passnet-solve","END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-solve\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":18},"01M799NNY9BKSM21V5R7QVTBJX","passnet-skill","PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under .\u002Fpass_dir\u002F, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-skill\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":18},"01M799NHH6NP7Z1RK77KGXAN6S","passnet-pattern-fusion","Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()\u002Freplacement_args()\u002Freplacement_func() files, when a pass \"failed to match\", or when deciding how to fuse multiple kernels\u002Fops.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-pattern-fusion\u002FSKILL.md",[],[],{"skillNo":101,"name":102,"nameZh":6,"description":103,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":18},"01M799ND2J4WNC2PGAZ0A06NDF","passnet-orchestrate","Round\u002Fbatch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep\u002Frevert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-orchestrate\u002FSKILL.md",[],[],{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":111,"categories":112,"keywords":113,"publishedAt":18},"01M799N8SBTMY02ZDG2AZF5ZWE","passnet-feedback","Fast, deterministic feedback for PassNet work: pre-flight pattern\u002Fmatch verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).","https:\u002F\u002Fgithub.com\u002FVectorSpaceLab\u002FAREX-Skill\u002Fblob\u002Fmain\u002Fskills\u002Ftask-oriented\u002FPassNet\u002Fskills\u002Fpassnet-feedback\u002FSKILL.md",[],[]]