[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EMXYP38HCD976JJ9WF8BSP":3,"skill-same:01M7EMXYP38HCD976JJ9WF8BSP":45},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M7EMXYP38HCD976JJ9WF8BSP","chaos-engine",null,"Canonical provider-neutral skill router and working contract. Use at the start of every task, on every host, in every main thread and delegate, before discovery, planning, edits, or answering.","MahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice","MahmoudElSharkawy","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F46620378?v=4","01M7EMXGM1HRD5QZ49MSFYKGRZ","Mahmoud ElSharkawy","01M7EMXGMKT9WCXGVQ82DZ9MY9",0,"MIT","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002F.chaos-engine\u002Fskills\u002Fchaos-engine\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: chaos-engine\ndescription: >-\n  Canonical provider-neutral skill router and working contract. Use at the start\n  of every task, on every host, in every main thread and delegate, before\n  discovery, planning, edits, or answering.\nlicense: MIT\n---\n\n# ChaosEngine\n\nCore card for every host and main thread; delegates load the\n[delegate card](..\u002F..\u002Freferences\u002Fdelegate-card.md). Adapters never restate\npolicy. Identity: [identity](..\u002F..\u002Fidentity.md).\n\n## Iron laws\n\n1. Measure thrice, cut once: research and plan before the first edit; depth\n   follows triage, ordering never changes.\n2. Evidence over inference. Inspect or run before claiming.\n3. Never weaken, delete, or rewrite a test to reach green. New behavior starts\n   with a red test; if test and requirement disagree, stop and say which is wrong.\n4. Smallest correct change, industry-standard output. Complete implementation\n   before its consolidated Check phase.\n5. Never claim a check you did not run.\n6. Finish before starting: pull-based flow, WIP limits, explicit Definition of\n   Done ([kanban](..\u002Fkanban\u002FSKILL.md)). Every finding ends fixed in this\n   delivery or filed as an issue; filing is allowed only when it is outside\n   scope and not critical, blocker, or high severity. Nothing stays only in\n   chat, a report, or a local file.\n7. Safety and ethics control: preserve user work and secrets; irreversible or\n   externally visible actions need explicit owner authority.\n\n## Triage\n\nAnswer in one line each before task-specific discovery:\n\n- **Blast radius**: one file, one module, or a public contract and its callers.\n- **Reversibility**: undone by deleting the diff, or touches persisted data, a\n  published artifact, or an external system?\n\n| Worse answer | Depth |\n| --- | --- |\n| One file, reversible | Read, fix, prove. No plan document. |\n| One module, reversible | Short plan: premises, two options, chosen one, proof. |\n| Public contract or hard to reverse | Written spec, owner decisions asked, full proof. |\n\nRe-triage when a premise is false. Before a broad search, run one\n`python3 .chaos-engine\u002Ftool.py retrieve` and record `skipped(\u003Creason>)`\n([retrieve-first](..\u002F..\u002Freferences\u002Fretrieve-first.md)). Gated: PR `## Plan` +\n`retrieve:` lines; `checkpoint --spec`.\n\n## Plan, then go\n\nAsk only plan-changing questions sources cannot answer. Many items: [delivery ledger](..\u002F..\u002Freferences\u002Fbot-entry.md#delivery-ledger) first (WIP limit). Once the plan is approved, go\nunattended to Done; stop only for a genuine owner decision or when a new\nrequest contradicts the plan.\n\n## Implement\n\nCompanions at task start via `tool.py entry` (hookless bots:\n[bot entry](..\u002F..\u002Freferences\u002Fbot-entry.md)): [Caveman ultra](..\u002F..\u002Fcompanions\u002Fcaveman-ultra.md),\n[Ponytail ultra](..\u002F..\u002Fcompanions\u002Fponytail-ultra.md).\nOff only: `stop caveman`, `stop ponytail`, `normal mode`. Vendor bodies only on\nexplicit invocation. Precedence: owner in chat > iron laws > this card >\nreferences > skills; deeper file wins on scope.\n\n## Check and deliver\n\n- Finish the implementation, then one Check: new or edited tests plus directly\n  impacted tests.\n- One fresh-context review after implementation; a second round only for\n  blocker findings.\n- One blocking CI wait per push (one background babysit loop to merged);\n  retry at most once. Stop and report.\n- Status reports are always a Markdown table: deltas, at most 15 lines.\n  `tool.py usage`.\n- Learning Session after every final delivery. Never local queues.\n- Repository work is not done until the pull request is merged.\n\n## Route\n\n\u003C!-- HARNESS-ROUTES:START -->\n| Route | Use when | Load |\n| --- | --- | --- |\n| Local LLM | local runtime or agents | [local-runtimes](..\u002Flocal-runtimes\u002FSKILL.md), [local-agency](..\u002Flocal-agency\u002FSKILL.md), [omniroute](..\u002Fomniroute\u002FSKILL.md), [freetoken](..\u002Ffreetoken\u002FSKILL.md), [colibri](..\u002Fcolibri\u002FSKILL.md), [local-openai-compat](..\u002Flocal-openai-compat\u002FSKILL.md), [local-coding-delegate](..\u002Flocal-coding-delegate\u002FSKILL.md) |\n| work-item | open or rewrite an issue or work item | [SKILL.md](..\u002Fwork-item\u002FSKILL.md) |\n| self-improve | after final delivery | [SKILL.md](..\u002Fself-improve\u002FSKILL.md) |\n| kanban | several deliverables, tickets, or delegates | [SKILL.md](..\u002Fkanban\u002FSKILL.md) |\n| architect | design, API, risk, review | [SKILL.md](..\u002Farchitect\u002FSKILL.md) |\n| Git cleanup | dirty worktree or stray branches | [SKILL.md](..\u002Fgit-cleanup\u002FSKILL.md) |\n| CI | slow merges, CI | [SKILL.md](..\u002Fci-cd\u002FSKILL.md) |\n| Zero-LLM first | install, doctor, repair by script | [zero-llm-catalog.md](..\u002F..\u002Freferences\u002Fzero-llm-catalog.md) |\n| Heal | drifted or unhealthy install | [heal-route.md](..\u002F..\u002Freferences\u002Fheal-route.md) |\n| Level-1 catalog | secondary skill or tool needed | [level-1-catalog.md](..\u002F..\u002Freferences\u002Flevel-1-catalog.md) |\n| Context firewall | isolate broad research | [context-firewall.md](..\u002F..\u002Freferences\u002Fcontext-firewall.md) |\n| Harness learn | traces show the overlay should change | [harness-learn.md](..\u002F..\u002Freferences\u002Fharness-learn.md) |\n| Design loop | design doc needs review rounds | [design-loop.md](..\u002F..\u002Freferences\u002Fdesign-loop.md) |\n| Deep research | cited multi-source research | [deep-research.md](..\u002F..\u002Freferences\u002Fdeep-research.md) |\n| UI delivery | user-visible UI | [ui-delivery.md](..\u002F..\u002Freferences\u002Fui-delivery.md) |\n| Public landing | calling card or product page | [public-landing.md](..\u002F..\u002Freferences\u002Fpublic-landing.md) |\n| Learn traces | turn traces into lessons | [learn-traces.md](..\u002F..\u002Freferences\u002Flearn-traces.md) |\n| Meta-optimize | periodic offline log review | [meta-optimize.md](..\u002F..\u002Freferences\u002Fmeta-optimize.md) |\n| Draft skill PR | opt-in eval-gated skill PR | [draft-skill-pr.md](..\u002F..\u002Freferences\u002Fdraft-skill-pr.md) |\n| Token budget | lean, balanced, or deep | [token-budget-modes.md](..\u002F..\u002Freferences\u002Ftoken-budget-modes.md) |\n| Eliminate waste | a hop or retry adds no decision | [eliminate-waste.md](..\u002F..\u002Freferences\u002Feliminate-waste.md) |\n| Prefer CLI over MCP | CLI and MCP both fit | [prefer-cli-over-mcp.md](..\u002F..\u002Freferences\u002Fprefer-cli-over-mcp.md) |\n| No proxy | a task would add a traffic proxy | [no-proxy.md](..\u002F..\u002Freferences\u002Fno-proxy.md) |\n| GAP-EXIT2 UX | host ignores exit-2 hard blocks | [host-parity-matrix.md](..\u002F..\u002Freferences\u002Fhost-parity-matrix.md) |\n| Add-ons | design, video, or project pack | [addons.md](..\u002F..\u002Freferences\u002Faddons.md) |\n| Complexity gate | complexity gate | [complexity-gate.md](..\u002F..\u002Freferences\u002Fcomplexity-gate.md) |\n| Durable jobs | outlives the session | [durable-jobs.md](..\u002F..\u002Freferences\u002Fdurable-jobs.md) |\n\u003C!-- HARNESS-ROUTES:END -->\n\n## Catalog\n\nOn demand: [catalog](..\u002F..\u002Freferences\u002Fcatalog.md),\n[router contract](..\u002F..\u002Freferences\u002Frouter-contract.md),\n[context economy](..\u002F..\u002Freferences\u002Fcontext-economy.md),\n[retrieve-first](..\u002F..\u002Freferences\u002Fretrieve-first.md),\n[script first](..\u002F..\u002Freferences\u002Fscript-first.md).\n","# ChaosEngine\n\nCore card for every host and main thread; delegates load the\n[delegate card](..\u002F..\u002Freferences\u002Fdelegate-card.md). Adapters never restate\npolicy. Identity: [identity](..\u002F..\u002Fidentity.md).\n\n## Iron laws\n\n1. Measure thrice, cut once: research and plan before the first edit; depth\n   follows triage, ordering never changes.\n2. Evidence over inference. Inspect or run before claiming.\n3. Never weaken, delete, or rewrite a test to reach green. New behavior starts\n   with a red test; if test and requirement disagree, stop and say which is wrong.\n4. Smallest correct change, industry-standard output. Complete implementation\n   before its consolidated Check phase.\n5. Never claim a check you did not run.\n6. Finish before starting: pull-based flow, WIP limits, explicit Definition of\n   Done ([kanban](..\u002Fkanban\u002FSKILL.md)). Every finding ends fixed in this\n   delivery or filed as an issue; filing is allowed only when it is outside\n   scope and not critical, blocker, or high severity. Nothing stays only in\n   chat, a report, or a local file.\n7. Safety and ethics control: preserve user work and secrets; irreversible or\n   externally visible actions need explicit owner authority.\n\n## Triage\n\nAnswer in one line each before task-specific discovery:\n\n- **Blast radius**: one file, one module, or a public contract and its callers.\n- **Reversibility**: undone by deleting the diff, or touches persisted data, a\n  published artifact, or an external system?\n\n| Worse answer | Depth |\n| --- | --- |\n| One file, reversible | Read, fix, prove. No plan document. |\n| One module, reversible | Short plan: premises, two options, chosen one, proof. |\n| Public contract or hard to reverse | Written spec, owner decisions asked, full proof. |\n\nRe-triage when a premise is false. Before a broad search, run one\n`python3 .chaos-engine\u002Ftool.py retrieve` and record `skipped(\u003Creason>)`\n([retrieve-first](..\u002F..\u002Freferences\u002Fretrieve-first.md)). Gated: PR `## Plan` +\n`retrieve:` lines; `checkpoint --spec`.\n\n## Plan, then go\n\nAsk only plan-changing questions sources cannot answer. Many items: [delivery ledger](..\u002F..\u002Freferences\u002Fbot-entry.md#delivery-ledger) first (WIP limit). Once the plan is approved, go\nunattended to Done; stop only for a genuine owner decision or when a new\nrequest contradicts the plan.\n\n## Implement\n\nCompanions at task start via `tool.py entry` (hookless bots:\n[bot entry](..\u002F..\u002Freferences\u002Fbot-entry.md)): [Caveman ultra](..\u002F..\u002Fcompanions\u002Fcaveman-ultra.md),\n[Ponytail ultra](..\u002F..\u002Fcompanions\u002Fponytail-ultra.md).\nOff only: `stop caveman`, `stop ponytail`, `normal mode`. Vendor bodies only on\nexplicit invocation. Precedence: owner in chat > iron laws > this card >\nreferences > skills; deeper file wins on scope.\n\n## Check and deliver\n\n- Finish the implementation, then one Check: new or edited tests plus directly\n  impacted tests.\n- One fresh-context review after implementation; a second round only for\n  blocker findings.\n- One blocking CI wait per push (one background babysit loop to merged);\n  retry at most once. Stop and report.\n- Status reports are always a Markdown table: deltas, at most 15 lines.\n  `tool.py usage`.\n- Learning Session after every final delivery. Never local queues.\n- Repository work is not done until the pull request is merged.\n\n## Route\n\n\u003C!-- HARNESS-ROUTES:START -->\n| Route | Use when | Load |\n| --- | --- | --- |\n| Local LLM | local runtime or agents | [local-runtimes](..\u002Flocal-runtimes\u002FSKILL.md), [local-agency](..\u002Flocal-agency\u002FSKILL.md), [omniroute](..\u002Fomniroute\u002FSKILL.md), [freetoken](..\u002Ffreetoken\u002FSKILL.md), [colibri](..\u002Fcolibri\u002FSKILL.md), [local-openai-compat](..\u002Flocal-openai-compat\u002FSKILL.md), [local-coding-delegate](..\u002Flocal-coding-delegate\u002FSKILL.md) |\n| work-item | open or rewrite an issue or work item | [SKILL.md](..\u002Fwork-item\u002FSKILL.md) |\n| self-improve | after final delivery | [SKILL.md](..\u002Fself-improve\u002FSKILL.md) |\n| kanban | several deliverables, tickets, or delegates | [SKILL.md](..\u002Fkanban\u002FSKILL.md) |\n| architect | design, API, risk, review | [SKILL.md](..\u002Farchitect\u002FSKILL.md) |\n| Git cleanup | dirty worktree or stray branches | [SKILL.md](..\u002Fgit-cleanup\u002FSKILL.md) |\n| CI | slow merges, CI | [SKILL.md](..\u002Fci-cd\u002FSKILL.md) |\n| Zero-LLM first | install, doctor, repair by script | [zero-llm-catalog.md](..\u002F..\u002Freferences\u002Fzero-llm-catalog.md) |\n| Heal | drifted or unhealthy install | [heal-route.md](..\u002F..\u002Freferences\u002Fheal-route.md) |\n| Level-1 catalog | secondary skill or tool needed | [level-1-catalog.md](..\u002F..\u002Freferences\u002Flevel-1-catalog.md) |\n| Context firewall | isolate broad research | [context-firewall.md](..\u002F..\u002Freferences\u002Fcontext-firewall.md) |\n| Harness learn | traces show the overlay should change | [harness-learn.md](..\u002F..\u002Freferences\u002Fharness-learn.md) |\n| Design loop | design doc needs review rounds | [design-loop.md](..\u002F..\u002Freferences\u002Fdesign-loop.md) |\n| Deep research | cited multi-source research | [deep-research.md](..\u002F..\u002Freferences\u002Fdeep-research.md) |\n| UI delivery | user-visible UI | [ui-delivery.md](..\u002F..\u002Freferences\u002Fui-delivery.md) |\n| Public landing | calling card or product page | [public-landing.md](..\u002F..\u002Freferences\u002Fpublic-landing.md) |\n| Learn traces | turn traces into lessons | [learn-traces.md](..\u002F..\u002Freferences\u002Flearn-traces.md) |\n| Meta-optimize | periodic offline log review | [meta-optimize.md](..\u002F..\u002Freferences\u002Fmeta-optimize.md) |\n| Draft skill PR | opt-in eval-gated skill PR | [draft-skill-pr.md](..\u002F..\u002Freferences\u002Fdraft-skill-pr.md) |\n| Token budget | lean, balanced, or deep | [token-budget-modes.md](..\u002F..\u002Freferences\u002Ftoken-budget-modes.md) |\n| Eliminate waste | a hop or retry adds no decision | [eliminate-waste.md](..\u002F..\u002Freferences\u002Feliminate-waste.md) |\n| Prefer CLI over MCP | CLI and MCP both fit | [prefer-cli-over-mcp.md](..\u002F..\u002Freferences\u002Fprefer-cli-over-mcp.md) |\n| No proxy | a task would add a traffic proxy | [no-proxy.md](..\u002F..\u002Freferences\u002Fno-proxy.md) |\n| GAP-EXIT2 UX | host ignores exit-2 hard blocks | [host-parity-matrix.md](..\u002F..\u002Freferences\u002Fhost-parity-matrix.md) |\n| Add-ons | design, video, or project pack | [addons.md](..\u002F..\u002Freferences\u002Faddons.md) |\n| Complexity gate | complexity gate | [complexity-gate.md](..\u002F..\u002Freferences\u002Fcomplexity-gate.md) |\n| Durable jobs | outlives the session | [durable-jobs.md](..\u002F..\u002Freferences\u002Fdurable-jobs.md) |\n\u003C!-- HARNESS-ROUTES:END -->\n\n## Catalog\n\nOn demand: [catalog](..\u002F..\u002Freferences\u002Fcatalog.md),\n[router contract](..\u002F..\u002Freferences\u002Frouter-contract.md),\n[context economy](..\u002F..\u002Freferences\u002Fcontext-economy.md),\n[retrieve-first](..\u002F..\u002Freferences\u002Fretrieve-first.md),\n[script first](..\u002F..\u002Freferences\u002Fscript-first.md).\n",{"name":5,"license":15,"description":7},"main","3456be4beee981fd5a2425c1f451e867daa99bb6",".chaos-engine\u002Fskills\u002Fchaos-engine","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy","SDET","2026-10-20 21:55:53",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add MahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice --skill chaos-engine",{"tool":37,"command":38},"bunx","bunx skills add MahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice --skill chaos-engine",{"tool":40,"command":41},"pnpm","pnpm dlx skills add MahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice --skill chaos-engine","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002F.chaos-engine\u002Fskills\u002Fchaos-engine\u002FSKILL.md",true,[],[46,53,60,69,76,83],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M7EMZSCKD5V0JPRVYD90AN95","ponytail","Lazy senior dev mode: the smallest change that fully solves the task, and a reply a busy human understands in one read. Use on any coding task (writing, fixing, refactoring, reviewing, choosing dependencies) and when the user says \"ponytail\", \"be lazy\", \"simplest solution\", \"yagni\", or complains about over-engineering or bloat. Levels: lite, full (default), ultra.","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002Fplugins\u002Fponytail\u002Fskills\u002Fponytail\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M7EMZ672QDBR1E5FM2D75NKE","work-item","Use when opening or rewriting a work item on any git-based SCM. Source-control agnostic; GitHub, GitLab, and Azure Boards are adapters only.","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002F.chaos-engine\u002Fskills\u002Fwork-item\u002FSKILL.md",[],[],{"skillNo":61,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":64,"licenseFileUrl":65,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M7EMZ305HDTCCW5VK62RQYAH","self-improve","Use when running ChaosEngine Learning Session self-improve. Harness lessons are GitHub issues only. Product lessons may queue after delivery or on request.","CC-BY-4.0","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002F.chaos-engine\u002Fskills\u002Fself-improve\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002F.chaos-engine\u002Fskills\u002Fself-improve\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7EMYYSCVHWETFSZJ7WB2GVY","omniroute","Use when local-runtimes selected the OmniRoute process for bounded work: runner, delegate continuity, capability enforcement and proof of dispatch.","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002F.chaos-engine\u002Fskills\u002Fomniroute\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7EMYW1RS1JNJ830EDXZNA8Z","local-runtimes","Use when choosing an optional local inference runtime for a narrow mechanical or offline job.","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002F.chaos-engine\u002Fskills\u002Flocal-runtimes\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7EMYRWMH5HBRR01KPK232BA","local-openai-compat","Loopback OpenAI-compatible peers (Ollama, LM Studio, llamacpp on 127.0.0.1). Use when a narrow task should go to an already READY loopback server.","https:\u002F\u002Fgithub.com\u002FMahmoudElSharkawy\u002FPlaywright-HarnessPOM-Practice\u002Fblob\u002Fmain\u002F.chaos-engine\u002Fskills\u002Flocal-openai-compat\u002FSKILL.md",[],[]]