[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7572QBZB4162KCDHBDEW4S3":3,"skill-same:01M7572QBZB4162KCDHBDEW4S3":46},{"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":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M7572QBZB4162KCDHBDEW4S3","pk-dynamic-workflows",null,"This skill should be used when the user wants to spawn subagents with different models, configure heterogeneous candidate backends, implement dynamic model routing, or run multi-model fusion workflows. Covers mock, local, premium API (Anthropic\u002FOpenAI), budget API (Kimi\u002FMiniMax\u002FQwen), routing layers (9Router), and CLI backends.","kingkillery\u002Foh-my-pk","kingkillery","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F200727508?v=4","01M6X59E9Q2FCEGBVQ1P6FT0YC","PK","01M756YT55MBWAW7Q3BJ768ADB",1,"MIT","https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk\u002Fblob\u002Fmain\u002Fpackages\u002Fverifier-extension\u002Fskills\u002Fpk-dynamic-workflows\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: pk-dynamic-workflows\ndescription: This skill should be used when the user wants to spawn subagents with different models, configure heterogeneous candidate backends, implement dynamic model routing, or run multi-model fusion workflows. Covers mock, local, premium API (Anthropic\u002FOpenAI), budget API (Kimi\u002FMiniMax\u002FQwen), routing layers (9Router), and CLI backends.\n---\n\n# Dynamic Subagent Workflows\nReference: `docs\u002Fdynamic-workflow-execution-map.md` maps backend\u002Fmodel options, router profiles, parallel candidate execution, and final synthesis.\n\n\n## Purpose\n\nUse this skill to spawn subagents with different AI models, configure heterogeneous backends, and implement dynamic routing for multi-model fusion workflows.\n\nPrefer this workflow when:\n- You want multiple candidates with different models for the same task\n- You need cost-aware backend selection (budget vs premium)\n- You want automatic fallback across providers\n- You're building a model ensemble or tournament\n- You want to compare outputs across model families\n\n## Available backends\n\n| Tier | Backend | Model | Cost (per M tok) | Use Case |\n|------|---------|-------|------------------|----------|\n| Mock | `mock` | `mock` | $0 | Tests, CI, validation |\n| Mock | `local` | `default` | $0 | Local testing alias |\n| Premium | `anthropic_api` | `claude-opus-4-8` | $5 \u002F $25 | Complex reasoning |\n| Premium | `openai_api` | `gpt-5.5` | $0.6 \u002F $2.5 | Synthesis |\n| Budget | `kimi` | `kimi-for-coding` | $0.6 \u002F $2.5 | Coding tasks |\n| Budget | `minimax` | `MiniMax-M3` | $0.3 \u002F $1.2 | Reasoning, cheapest |\n| Budget | `qwen` | `qwen-coder-plus` | ~$0.3 \u002F $0.6 | Strong coding, Chinese\u002FEnglish |\n| Router | `9router` | Any (60+ providers) | Pass-through | Auto-fallback, quota pooling |\n| CLI | `subprocess_cli` | Operator-configured | Local compute | Custom agent binaries |\n\n## Quick start: Spawn multiple candidates\n\n```python\nfrom harness.core.lifecycle import BACKENDS\nfrom harness.agents.base import AgentRunRequest\nfrom harness.core.task_contract import load_task_contract\n\ntask = load_task_contract(Path(\"my_task.json\"))\n\n# Spawn 3 candidates with different models\nbackends = {\n    \"minimax\": BACKENDS[\"minimax\"],      # Lowest cost\n    \"qwen\": BACKENDS[\"qwen\"],            # Strong code generation\n    \"openai_api\": BACKENDS[\"openai_api\"], # Best synthesis\n}\n\nresults = []\nfor backend_name, backend in backends.items():\n    request = AgentRunRequest(\n        run_id=\"run-1\",\n        candidate_id=f\"cand-{backend_name}\",\n        task_contract=task,\n        workspace_path=\"\u002Ftmp\u002Fworkspace\",\n        role=\"coder\",\n        prompt=\"Implement the feature\",\n        model=\"default\",  # each backend resolves its configured default\n        trace_path=f\"\u002Ftmp\u002Ftraces\u002F{backend_name}.json\",\n    )\n    results.append(backend.run(request))\n```\n\n## Model resolution order\n\nAll API backends follow the same resolution:\n\n1. **Request override:** `request.model` (if not \"default\" or \"mock\")\n2. **Environment override:** `FMH_*_MODEL` env var\n3. **Default fallback:** Hardcoded backend default\n\nExample: Set `FMH_QWEN_MODEL=qwen-max` to globally override Qwen's model.\n\n## Using 9Router for resilient runs\n\n9Router is a local gateway that routes to 60+ providers with auto-fallback:\n\n```bash\n# Install and start\nnpm install -g 9router\n9router  # Starts localhost:20128\u002Fv1\n```\n\n```python\n# Use in harness — no API keys needed locally\nbackend = BACKENDS[\"9router\"]\nrequest = AgentRunRequest(\n    ...,\n    model=\"qwen-coder-plus\",  # Any 9Router-supported model\n)\nresult = backend.run(request)\n```\n\nFeatures:\n- 3-tier fallback: subscription → cheap → free\n- RTK token saver: -20–40% on tool results\n- Real-time quota dashboard\n- Multi-account per provider\n\n## Dynamic backend selection\n\n```python\ndef pick_backend(task):\n    \"\"\"Select backend based on task properties and budget.\"\"\"\n    if task.budget.max_total_usd \u003C 0.5:\n        return BACKENDS[\"minimax\"]  # Cheapest\n    if \"chinese\" in task.user_request.lower():\n        return BACKENDS[\"qwen\"]     # Chinese-optimized\n    if task.fusion.candidate_count > 5:\n        return BACKENDS[\"9router\"]  # Quota pooling\n    return BACKENDS[\"openai_api\"]   # Default premium\n\nbackend = pick_backend(task)\n```\n\n## Dynamic router profile\n\nUse `StaticRouter(profile=\"dynamic\")` when the harness should create a mixed candidate slate automatically:\n\n```python\nfrom harness.routing.router import StaticRouter\n\ndecision = StaticRouter(profile=\"dynamic\").route(task, backend=\"mock\")\nassert [plan.backend for plan in decision.candidates] == [\"qwen\", \"minimax\", \"kimi\"]  # for 3 candidates\n```\n\nFor up to 5 candidates, the dynamic pool is:\n\n1. `qwen` — coding-oriented candidate\n2. `minimax` — cheapest reasoning candidate\n3. `kimi` — alternate budget coding\u002Frequirements lane\n4. `9router` — resilient fallback\u002Fquota pooling lane\n5. `openai_api` — premium synthesis\u002Fgeneralist lane\n\nEvery dynamic candidate uses `model=\"default\"`; each backend resolves its own configured default at execution time.\n\n## Environment variables\n\n| Backend | Required | Optional override |\n|---------|----------|-------------------|\n| `anthropic_api` | `ANTHROPIC_API_KEY` | `FMH_ANTHROPIC_MODEL` |\n| `openai_api` | `OPENAI_API_KEY` | `FMH_OPENAI_MODEL` |\n| `kimi` | `KIMI_API_KEY` or `MOONSHOT_API_KEY` | `FMH_KIMI_MODEL` |\n| `minimax` | `MINIMAX_API_KEY` | `FMH_MINIMAX_MODEL` |\n| `qwen` | `DASHSCOPE_API_KEY` | `FMH_QWEN_MODEL` |\n| `9router` | None (local) or `9ROUTER_API_KEY` (cloud) | `FMH_9ROUTER_MODEL` |\n| `subprocess_cli` | `FMH_SUBPROCESS_CLI_CMD` | — |\n\n## Fail-closed behavior\n\nAPI backends fail closed when required credentials are missing. 9Router is the exception for local `http:\u002F\u002Flocalhost:20128\u002Fv1`: it uses a harmless `local-9router` bearer token when `9ROUTER_API_KEY` is unset, while remote\u002Fcloud 9Router URLs still require an explicit key.\n\n## Structured output contract\n\nAll API backends enforce the same JSON schema:\n\n```json\n{\n  \"answer\": \"string\",\n  \"confidence\": 0.0-1.0,\n  \"evidence\": [{\"type\": \"...\", \"source\": \"...\", \"claim\": \"...\", \"confidence\": 0.0-1.0}],\n  \"self_assessment\": {\"confidence\": 0.0-1.0, \"assumptions\": [\"...\"]},\n  \"metrics\": {\"latency_ms\": 0, \"cost_usd\": 0.0, \"tool_calls\": 0}\n}\n```\n\nThe parser tolerates markdown fences, prose wrappers, and `\u003Cthink>` blocks.\n\n## Smoke test command\n\nRun the reusable smoke script after changing backend registration, model resolution, 9Router auth, or dynamic routing:\n\n```bash\npython scripts\u002Ftest_dynamic_workflows.py\n```\n\nIt verifies:\n- all expected backends are registered,\n- Qwen and 9Router model resolution works,\n- local 9Router auth works without `9ROUTER_API_KEY`,\n- mock candidate execution completes,\n- `profile=\"dynamic\"` emits the expected 3-candidate and 5-candidate slates.\n\n## Adding a new backend\n\n1. Create class in `harness\u002Fagents\u002F` inheriting `OpenAICompatibleBackend` or `AnthropicCompatibleBackend`\n2. Set `config` with label, API key envs, base URL, default model, pricing\n3. Add entry to `configs\u002Fmodels.yaml`\n4. Register in `harness\u002Fcore\u002Flifecycle.py` `BACKENDS` dict\n5. Add tests in `tests\u002Funit\u002Ftest_backends.py`\n\n## Rules\n\n- Always use `request.model` to control backend selection per candidate\n- Prefer `9router` for production resilience (auto-fallback)\n- Use `mock` only for tests and pipeline validation\n- Set `FMH_*_MODEL` env vars for global model overrides\n- Gather deterministic evidence before comparing candidate outputs\n- Keep candidate payloads focused; trim unrelated noise\n- Treat test results as stronger evidence than model self-assessment\n\n## Success criteria\n\nThis skill is successful when it:\n- Spawns multiple candidates with different models\n- Selects backends based on task properties and budget\n- Handles missing credentials gracefully (fail-closed)\n- Produces comparable structured outputs across model families\n- Enables resilient production runs via routing layers\n","# Dynamic Subagent Workflows\nReference: `docs\u002Fdynamic-workflow-execution-map.md` maps backend\u002Fmodel options, router profiles, parallel candidate execution, and final synthesis.\n\n\n## Purpose\n\nUse this skill to spawn subagents with different AI models, configure heterogeneous backends, and implement dynamic routing for multi-model fusion workflows.\n\nPrefer this workflow when:\n- You want multiple candidates with different models for the same task\n- You need cost-aware backend selection (budget vs premium)\n- You want automatic fallback across providers\n- You're building a model ensemble or tournament\n- You want to compare outputs across model families\n\n## Available backends\n\n| Tier | Backend | Model | Cost (per M tok) | Use Case |\n|------|---------|-------|------------------|----------|\n| Mock | `mock` | `mock` | $0 | Tests, CI, validation |\n| Mock | `local` | `default` | $0 | Local testing alias |\n| Premium | `anthropic_api` | `claude-opus-4-8` | $5 \u002F $25 | Complex reasoning |\n| Premium | `openai_api` | `gpt-5.5` | $0.6 \u002F $2.5 | Synthesis |\n| Budget | `kimi` | `kimi-for-coding` | $0.6 \u002F $2.5 | Coding tasks |\n| Budget | `minimax` | `MiniMax-M3` | $0.3 \u002F $1.2 | Reasoning, cheapest |\n| Budget | `qwen` | `qwen-coder-plus` | ~$0.3 \u002F $0.6 | Strong coding, Chinese\u002FEnglish |\n| Router | `9router` | Any (60+ providers) | Pass-through | Auto-fallback, quota pooling |\n| CLI | `subprocess_cli` | Operator-configured | Local compute | Custom agent binaries |\n\n## Quick start: Spawn multiple candidates\n\n```python\nfrom harness.core.lifecycle import BACKENDS\nfrom harness.agents.base import AgentRunRequest\nfrom harness.core.task_contract import load_task_contract\n\ntask = load_task_contract(Path(\"my_task.json\"))\n\n# Spawn 3 candidates with different models\nbackends = {\n    \"minimax\": BACKENDS[\"minimax\"],      # Lowest cost\n    \"qwen\": BACKENDS[\"qwen\"],            # Strong code generation\n    \"openai_api\": BACKENDS[\"openai_api\"], # Best synthesis\n}\n\nresults = []\nfor backend_name, backend in backends.items():\n    request = AgentRunRequest(\n        run_id=\"run-1\",\n        candidate_id=f\"cand-{backend_name}\",\n        task_contract=task,\n        workspace_path=\"\u002Ftmp\u002Fworkspace\",\n        role=\"coder\",\n        prompt=\"Implement the feature\",\n        model=\"default\",  # each backend resolves its configured default\n        trace_path=f\"\u002Ftmp\u002Ftraces\u002F{backend_name}.json\",\n    )\n    results.append(backend.run(request))\n```\n\n## Model resolution order\n\nAll API backends follow the same resolution:\n\n1. **Request override:** `request.model` (if not \"default\" or \"mock\")\n2. **Environment override:** `FMH_*_MODEL` env var\n3. **Default fallback:** Hardcoded backend default\n\nExample: Set `FMH_QWEN_MODEL=qwen-max` to globally override Qwen's model.\n\n## Using 9Router for resilient runs\n\n9Router is a local gateway that routes to 60+ providers with auto-fallback:\n\n```bash\n# Install and start\nnpm install -g 9router\n9router  # Starts localhost:20128\u002Fv1\n```\n\n```python\n# Use in harness — no API keys needed locally\nbackend = BACKENDS[\"9router\"]\nrequest = AgentRunRequest(\n    ...,\n    model=\"qwen-coder-plus\",  # Any 9Router-supported model\n)\nresult = backend.run(request)\n```\n\nFeatures:\n- 3-tier fallback: subscription → cheap → free\n- RTK token saver: -20–40% on tool results\n- Real-time quota dashboard\n- Multi-account per provider\n\n## Dynamic backend selection\n\n```python\ndef pick_backend(task):\n    \"\"\"Select backend based on task properties and budget.\"\"\"\n    if task.budget.max_total_usd \u003C 0.5:\n        return BACKENDS[\"minimax\"]  # Cheapest\n    if \"chinese\" in task.user_request.lower():\n        return BACKENDS[\"qwen\"]     # Chinese-optimized\n    if task.fusion.candidate_count > 5:\n        return BACKENDS[\"9router\"]  # Quota pooling\n    return BACKENDS[\"openai_api\"]   # Default premium\n\nbackend = pick_backend(task)\n```\n\n## Dynamic router profile\n\nUse `StaticRouter(profile=\"dynamic\")` when the harness should create a mixed candidate slate automatically:\n\n```python\nfrom harness.routing.router import StaticRouter\n\ndecision = StaticRouter(profile=\"dynamic\").route(task, backend=\"mock\")\nassert [plan.backend for plan in decision.candidates] == [\"qwen\", \"minimax\", \"kimi\"]  # for 3 candidates\n```\n\nFor up to 5 candidates, the dynamic pool is:\n\n1. `qwen` — coding-oriented candidate\n2. `minimax` — cheapest reasoning candidate\n3. `kimi` — alternate budget coding\u002Frequirements lane\n4. `9router` — resilient fallback\u002Fquota pooling lane\n5. `openai_api` — premium synthesis\u002Fgeneralist lane\n\nEvery dynamic candidate uses `model=\"default\"`; each backend resolves its own configured default at execution time.\n\n## Environment variables\n\n| Backend | Required | Optional override |\n|---------|----------|-------------------|\n| `anthropic_api` | `ANTHROPIC_API_KEY` | `FMH_ANTHROPIC_MODEL` |\n| `openai_api` | `OPENAI_API_KEY` | `FMH_OPENAI_MODEL` |\n| `kimi` | `KIMI_API_KEY` or `MOONSHOT_API_KEY` | `FMH_KIMI_MODEL` |\n| `minimax` | `MINIMAX_API_KEY` | `FMH_MINIMAX_MODEL` |\n| `qwen` | `DASHSCOPE_API_KEY` | `FMH_QWEN_MODEL` |\n| `9router` | None (local) or `9ROUTER_API_KEY` (cloud) | `FMH_9ROUTER_MODEL` |\n| `subprocess_cli` | `FMH_SUBPROCESS_CLI_CMD` | — |\n\n## Fail-closed behavior\n\nAPI backends fail closed when required credentials are missing. 9Router is the exception for local `http:\u002F\u002Flocalhost:20128\u002Fv1`: it uses a harmless `local-9router` bearer token when `9ROUTER_API_KEY` is unset, while remote\u002Fcloud 9Router URLs still require an explicit key.\n\n## Structured output contract\n\nAll API backends enforce the same JSON schema:\n\n```json\n{\n  \"answer\": \"string\",\n  \"confidence\": 0.0-1.0,\n  \"evidence\": [{\"type\": \"...\", \"source\": \"...\", \"claim\": \"...\", \"confidence\": 0.0-1.0}],\n  \"self_assessment\": {\"confidence\": 0.0-1.0, \"assumptions\": [\"...\"]},\n  \"metrics\": {\"latency_ms\": 0, \"cost_usd\": 0.0, \"tool_calls\": 0}\n}\n```\n\nThe parser tolerates markdown fences, prose wrappers, and `\u003Cthink>` blocks.\n\n## Smoke test command\n\nRun the reusable smoke script after changing backend registration, model resolution, 9Router auth, or dynamic routing:\n\n```bash\npython scripts\u002Ftest_dynamic_workflows.py\n```\n\nIt verifies:\n- all expected backends are registered,\n- Qwen and 9Router model resolution works,\n- local 9Router auth works without `9ROUTER_API_KEY`,\n- mock candidate execution completes,\n- `profile=\"dynamic\"` emits the expected 3-candidate and 5-candidate slates.\n\n## Adding a new backend\n\n1. Create class in `harness\u002Fagents\u002F` inheriting `OpenAICompatibleBackend` or `AnthropicCompatibleBackend`\n2. Set `config` with label, API key envs, base URL, default model, pricing\n3. Add entry to `configs\u002Fmodels.yaml`\n4. Register in `harness\u002Fcore\u002Flifecycle.py` `BACKENDS` dict\n5. Add tests in `tests\u002Funit\u002Ftest_backends.py`\n\n## Rules\n\n- Always use `request.model` to control backend selection per candidate\n- Prefer `9router` for production resilience (auto-fallback)\n- Use `mock` only for tests and pipeline validation\n- Set `FMH_*_MODEL` env vars for global model overrides\n- Gather deterministic evidence before comparing candidate outputs\n- Keep candidate payloads focused; trim unrelated noise\n- Treat test results as stronger evidence than model self-assessment\n\n## Success criteria\n\nThis skill is successful when it:\n- Spawns multiple candidates with different models\n- Selects backends based on task properties and budget\n- Handles missing credentials gracefully (fail-closed)\n- Produces comparable structured outputs across model families\n- Enables resilient production runs via routing layers\n",{"name":5,"description":7},"main","70395f201d2623f8048709ec3af66deb58ac4cfc","packages\u002Fverifier-extension\u002Fskills\u002Fpk-dynamic-workflows",0,"https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk","https:\u002F\u002Fomp.sh","https:\u002F\u002Fgithub.com\u002Fkingkillery","2026-11-14 17:53:41",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add kingkillery\u002Foh-my-pk --skill pk-dynamic-workflows",{"tool":38,"command":39},"bunx","bunx skills add kingkillery\u002Foh-my-pk --skill pk-dynamic-workflows",{"tool":41,"command":42},"pnpm","pnpm dlx skills add kingkillery\u002Foh-my-pk --skill pk-dynamic-workflows","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk\u002Fblob\u002Fmain\u002Fpackages\u002Fverifier-extension\u002Fskills\u002Fpk-dynamic-workflows\u002FSKILL.md",true,[],[47,54,60,65,72,79],{"skillNo":48,"name":49,"nameZh":6,"description":50,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":51,"categories":52,"keywords":53,"publishedAt":19},"01M7572XYPFKX774ZDDEF4A210","llm-as-verifier","This skill should be used when the user asks to \"compare candidate patches\", \"rank multiple solutions\", \"choose between implementations\", \"verify two drafts\", \"pick the best trajectory\", \"use llm-as-a-verifier\", or wants repeated criteria-decomposed LLM verification instead of a single free-form judgment.","https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk\u002Fblob\u002Fmain\u002Fpython\u002Ffugu\u002F.agents\u002Fskills\u002Fllm-as-verifier\u002FSKILL.md",[],[],{"skillNo":55,"name":56,"nameZh":6,"description":6,"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},"01M7572V10JTX07X4EV35RFEEW","pk-subagent-orchestrator","https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk\u002Fblob\u002Fmain\u002Fpackages\u002Fverifier-extension\u002Fskills\u002Fpk-subagent-orchestrator\u002FSKILL.md",[],[],{"skillNo":61,"name":49,"nameZh":6,"description":50,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"01M7572KGQKE6MCBAQJS713443","https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk\u002Fblob\u002Fmain\u002Fpackages\u002Fverifier-extension\u002Fskills\u002Fllm-as-verifier\u002FSKILL.md",[],[],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"01M7572FQHTHCGKWNA603A6209","fmh","Run fmh (fusion-meta-harness) — verifier fusion, evaluate-verifier, improve-verifier, promote, run-task. Orchestrates multi-agent candidate generation, rubric scoring, synthesis, and reliability gates for the pi-llm-as-verifier project.","https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk\u002Fblob\u002Fmain\u002Fpackages\u002Fverifier-extension\u002Fskills\u002Ffmh\u002FSKILL.md",[],[],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M7572EPWC4T7TCYHNTR7CEQT","greet","A greeting skill for testing","https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk\u002Fblob\u002Fmain\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Fmarketplace\u002Ffixtures\u002Fvalid-marketplace\u002Fplugins\u002Fhello-plugin\u002Fskills\u002Fgreet\u002FSKILL.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M7572D377961TGVQNQB7PFNH","valid-skill","A valid skill for testing purposes.","https:\u002F\u002Fgithub.com\u002Fkingkillery\u002Foh-my-pk\u002Fblob\u002Fmain\u002Fpackages\u002Fcoding-agent\u002Ftest\u002Ffixtures\u002Fskills\u002Fvalid-skill\u002FSKILL.md",[],[]]