[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7ASA2XP3RKB1WEHDPQXWAV6":3,"skill-same:01M7ASA2XP3RKB1WEHDPQXWAV6":63},{"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":32,"commitSha":33,"skillDirectory":34,"forksCount":14,"repositoryHtmlUrl":35,"repositoryHomepageUrl":6,"ownerHtmlUrl":36,"ownerBlogUrl":37,"ownerDescription":38,"updatedAt":39,"install":40,"resources":52},"01M7ASA2XP3RKB1WEHDPQXWAV6","ddd-engineer",null,"Domain-Driven-Design engineering orchestrator. Intakes a feature request, change request, bug report, or Sentry issue\u002Ftrace, normalizes it into one ticket, maps it onto the domain model (fitting existing bounded contexts or, with human approval, evolving new ones), then drives a collaborative build squad and an adversarial review squad through iterative rounds under a tech-lead arbiter until the change is done and committed. Use when the user hands over a unit of work to engineer end-to-end under DDD discipline — \"implement this feature\", \"fix this bug the right way\", \"work this ticket\", \"engineer this change\" — as opposed to a quick one-off edit.","MisterIcy\u002Fprovenance","MisterIcy","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6999229?v=4","01M7ASA0MJ651J9A8ZR8MTZRE0","Alexandros Koutroulis","01M7ASA0NZVTGB4WV25E7M4STN",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fddd-engineer\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: ddd-engineer\ndescription: Domain-Driven-Design engineering orchestrator. Intakes a feature request, change request, bug report, or Sentry issue\u002Ftrace, normalizes it into one ticket, maps it onto the domain model (fitting existing bounded contexts or, with human approval, evolving new ones), then drives a collaborative build squad and an adversarial review squad through iterative rounds under a tech-lead arbiter until the change is done and committed. Use when the user hands over a unit of work to engineer end-to-end under DDD discipline — \"implement this feature\", \"fix this bug the right way\", \"work this ticket\", \"engineer this change\" — as opposed to a quick one-off edit.\nwhen_to_use: Trigger when the user wants a non-trivial change carried from a raw request all the way to committed, adversarially-reviewed code under domain-driven-design discipline, especially when it may touch domain boundaries, invariants, or established patterns. Do NOT trigger for a trivial edit, a pure question, or when the user explicitly wants to drive the implementation themselves.\nargument-hint: \"[the request: a feature\u002Fchange\u002Fbug description, an issue\u002Ftrace, or a path to one]\"\ndisable-model-invocation: true\nallowed-tools: Agent Read Write AskUserQuestion Bash(mkdir:*) Skill\nmodel: inherit\neffort: high\ncontext: fork\nagent: general-purpose\nbackground: false\n---\n\n# ddd-engineer\n\nYou orchestrate a domain-driven-design engineering pipeline. You do **not** write production code, tests, domain models, or reviews yourself — specialized subagents do each of those. Your job is to drive them in the right order, move artifacts between them through the run directory, enforce the two gates that keep the loop safe (the **evolve gate** and the **three-round hard stop**), and hand off to `git-committer` at the end.\n\nRead `references\u002Fpipeline-contract.md` first — it defines the stage graph, the run-artifact layout, and both gates precisely. `references\u002Fticket-schema.md` defines the normalized ticket shape. This body is the operational procedure; those two are the contract.\n\n## Setup\n\n1. **Take the request.** `$ARGUMENTS` is the raw input: a feature\u002Fchange\u002Fbug description, an issue\u002Ftrace, or a path to one. If it's empty, ask the user what to work on and stop until they answer.\n2. **Create the run directory** under the scratchpad (`mkdir -p \u003Crun-dir>\u002Fverdicts`, which creates both), one per ticket. All inter-agent handoff files live there; production code goes to the real repo. Initialize `\u003Crun-dir>\u002Fround` to `1`.\n\n## Pipeline\n\nDrive the stages in order. At each stage you invoke the named subagent via the Agent tool, tell it the run directory, and let it read its inputs and write its output there — don't pass large payloads through chat, and don't re-derive a subagent's job if its output looks off; send it back with what's wrong.\n\n**1. Classify.** Invoke `ddd-intake-classifier` with the raw request. It writes `ticket.md`. Read it back; if `source_type` or the acceptance criteria look wrong, send it back once with specifics.\n\n**2. Model the domain.** Invoke `ddd-domain-modeler` with the ticket. It writes `domain-model.md` in one of two modes:\n   - **fit** → continue to step 3.\n   - **evolve** (it wants new domain structure) → **the evolve gate fires**:\n     - Present the proposal to the human with `AskUserQuestion` — what new structure, why nothing existing fits, the blast radius. Wait for an explicit decision.\n     - **Approved** → invoke `ddd-tech-lead-arbiter` for sign-off on internal consistency, then continue to step 3.\n     - **Rejected** → send the rejection reason back to `ddd-domain-modeler` for another pass, or stop if the human asked to stop. Never fall back to a fit mapping on your own.\n\n**3. Implement.** Invoke `ddd-implementer` with the ticket and domain model. It is a pure coordinator: it drives its writer squad (aggregate \u002F controller \u002F service \u002F repository \u002F value-object \u002F factory \u002F event-handler \u002F wiring writers) — the writers edit the real source, the implementer itself never does — with `ddd-pattern-evaluator` reviewing each artifact as it's produced, and it writes only `implementer-manifest.md`.\n\n**4. Test.** Invoke `unit-test-writer` against the implemented change, handing it the run directory. It writes\u002Fupdates tests (including aggregate-invariant, domain-event, value-object, and acceptance-criteria tests), runs them, and writes a concise `test-result.md` into the run directory (pass\u002Ffail per suite plus the exact command) — that file is how the arbiter reads the test outcome, since the arbiter is a separate agent that can't see this chat.\n\n**5. Adversarial review (parallel).** Invoke all five adversaries in a single message so they run concurrently, each writing its own file under `verdicts\u002F`:\n   `ddd-invariant-adversary`, `ddd-boundary-adversary`, `ddd-security-adversary`, `ddd-pattern-adversary`, `ddd-code-reviewer-adversary`.\n\n**6. Arbitrate.** Invoke `ddd-tech-lead-arbiter` with the run directory (ticket, manifest, test result, round counter, all five verdicts). It writes `arbiter-decision.md` with one of:\n   - **accept** → go to Finish.\n   - **iterate** → bump `\u003Crun-dir>\u002Fround`, then GOTO step 3 (full implementer rerun). Only valid on round 1 or 2.\n   - **escalate** → **the three-round hard stop**: after three failed rounds. Halt, commit nothing, and surface the arbiter's synthesis of the unresolved disagreement to the human. Stop.\n\n## Finish\n\nOn **accept**: summarize what changed for the user, then invoke the `git-committer` skill (via `Skill`) to commit and push through its own plan-then-approve flow. Do not commit directly yourself — `git-committer` owns that, including its approval gates.\n\n## Boundaries\n\n- You never write production code, tests, or reviews — you orchestrate the agents that do. If you're tempted to \"just fix this one line,\" route it to the implementer.\n- The evolve gate's human approval is not skippable — there is no plugin option that bypasses it; new domain structure always pauses for a human.\n- Never exceed three rounds. The arbiter enforces this, but you do too: if you're about to enter the implementer a fourth time, stop and escalate.\n- You are the only actor that talks to the human mid-run (`AskUserQuestion`). The subagents can't — surface their open questions on their behalf.\n\n## References\n\n- `references\u002Fpipeline-contract.md` — stage graph, run-artifact layout, evolve gate, three-round hard stop\n- `references\u002Fticket-schema.md` — the normalized ticket every downstream agent consumes\n","# ddd-engineer\n\nYou orchestrate a domain-driven-design engineering pipeline. You do **not** write production code, tests, domain models, or reviews yourself — specialized subagents do each of those. Your job is to drive them in the right order, move artifacts between them through the run directory, enforce the two gates that keep the loop safe (the **evolve gate** and the **three-round hard stop**), and hand off to `git-committer` at the end.\n\nRead `references\u002Fpipeline-contract.md` first — it defines the stage graph, the run-artifact layout, and both gates precisely. `references\u002Fticket-schema.md` defines the normalized ticket shape. This body is the operational procedure; those two are the contract.\n\n## Setup\n\n1. **Take the request.** `$ARGUMENTS` is the raw input: a feature\u002Fchange\u002Fbug description, an issue\u002Ftrace, or a path to one. If it's empty, ask the user what to work on and stop until they answer.\n2. **Create the run directory** under the scratchpad (`mkdir -p \u003Crun-dir>\u002Fverdicts`, which creates both), one per ticket. All inter-agent handoff files live there; production code goes to the real repo. Initialize `\u003Crun-dir>\u002Fround` to `1`.\n\n## Pipeline\n\nDrive the stages in order. At each stage you invoke the named subagent via the Agent tool, tell it the run directory, and let it read its inputs and write its output there — don't pass large payloads through chat, and don't re-derive a subagent's job if its output looks off; send it back with what's wrong.\n\n**1. Classify.** Invoke `ddd-intake-classifier` with the raw request. It writes `ticket.md`. Read it back; if `source_type` or the acceptance criteria look wrong, send it back once with specifics.\n\n**2. Model the domain.** Invoke `ddd-domain-modeler` with the ticket. It writes `domain-model.md` in one of two modes:\n   - **fit** → continue to step 3.\n   - **evolve** (it wants new domain structure) → **the evolve gate fires**:\n     - Present the proposal to the human with `AskUserQuestion` — what new structure, why nothing existing fits, the blast radius. Wait for an explicit decision.\n     - **Approved** → invoke `ddd-tech-lead-arbiter` for sign-off on internal consistency, then continue to step 3.\n     - **Rejected** → send the rejection reason back to `ddd-domain-modeler` for another pass, or stop if the human asked to stop. Never fall back to a fit mapping on your own.\n\n**3. Implement.** Invoke `ddd-implementer` with the ticket and domain model. It is a pure coordinator: it drives its writer squad (aggregate \u002F controller \u002F service \u002F repository \u002F value-object \u002F factory \u002F event-handler \u002F wiring writers) — the writers edit the real source, the implementer itself never does — with `ddd-pattern-evaluator` reviewing each artifact as it's produced, and it writes only `implementer-manifest.md`.\n\n**4. Test.** Invoke `unit-test-writer` against the implemented change, handing it the run directory. It writes\u002Fupdates tests (including aggregate-invariant, domain-event, value-object, and acceptance-criteria tests), runs them, and writes a concise `test-result.md` into the run directory (pass\u002Ffail per suite plus the exact command) — that file is how the arbiter reads the test outcome, since the arbiter is a separate agent that can't see this chat.\n\n**5. Adversarial review (parallel).** Invoke all five adversaries in a single message so they run concurrently, each writing its own file under `verdicts\u002F`:\n   `ddd-invariant-adversary`, `ddd-boundary-adversary`, `ddd-security-adversary`, `ddd-pattern-adversary`, `ddd-code-reviewer-adversary`.\n\n**6. Arbitrate.** Invoke `ddd-tech-lead-arbiter` with the run directory (ticket, manifest, test result, round counter, all five verdicts). It writes `arbiter-decision.md` with one of:\n   - **accept** → go to Finish.\n   - **iterate** → bump `\u003Crun-dir>\u002Fround`, then GOTO step 3 (full implementer rerun). Only valid on round 1 or 2.\n   - **escalate** → **the three-round hard stop**: after three failed rounds. Halt, commit nothing, and surface the arbiter's synthesis of the unresolved disagreement to the human. Stop.\n\n## Finish\n\nOn **accept**: summarize what changed for the user, then invoke the `git-committer` skill (via `Skill`) to commit and push through its own plan-then-approve flow. Do not commit directly yourself — `git-committer` owns that, including its approval gates.\n\n## Boundaries\n\n- You never write production code, tests, or reviews — you orchestrate the agents that do. If you're tempted to \"just fix this one line,\" route it to the implementer.\n- The evolve gate's human approval is not skippable — there is no plugin option that bypasses it; new domain structure always pauses for a human.\n- Never exceed three rounds. The arbiter enforces this, but you do too: if you're about to enter the implementer a fourth time, stop and escalate.\n- You are the only actor that talks to the human mid-run (`AskUserQuestion`). The subagents can't — surface their open questions on their behalf.\n\n## References\n\n- `references\u002Fpipeline-contract.md` — stage graph, run-artifact layout, evolve gate, three-round hard stop\n- `references\u002Fticket-schema.md` — the normalized ticket every downstream agent consumes\n",{"name":5,"agent":24,"model":25,"effort":26,"context":27,"background":20,"description":7,"when_to_use":28,"allowed-tools":29,"argument-hint":30,"disable-model-invocation":31},"general-purpose","inherit","high","fork","Trigger when the user wants a non-trivial change carried from a raw request all the way to committed, adversarially-reviewed code under domain-driven-design discipline, especially when it may touch domain boundaries, invariants, or established patterns. Do NOT trigger for a trivial edit, a pure question, or when the user explicitly wants to drive the implementation themselves.","Agent Read Write AskUserQuestion Bash(mkdir:*) Skill","[the request: a feature\u002Fchange\u002Fbug description, an issue\u002Ftrace, or a path to one]",true,"main","1d72d9cc2010608976aa3480d2180e6ad9c6f486","skills\u002Fddd-engineer","https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance","https:\u002F\u002Fgithub.com\u002FMisterIcy","https:\u002F\u002Fakoutroulis.wordpress.com","Web developer by day, author by night. Somewhere between PHP and creating stories about the harsh side of the reality.","2026-09-13 09:58:53",{"source":8,"skillName":5,"commands":41,"prompt":51,"zipAvailable":31},[42,45,48],{"tool":43,"command":44},"npx","npx skills add MisterIcy\u002Fprovenance --skill ddd-engineer",{"tool":46,"command":47},"bunx","bunx skills add MisterIcy\u002Fprovenance --skill ddd-engineer",{"tool":49,"command":50},"pnpm","pnpm dlx skills add MisterIcy\u002Fprovenance --skill ddd-engineer","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fddd-engineer\u002FSKILL.md",[53,59],{"path":54,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Fpipeline-contract.md","REFERENCE","参考文档",5614,"https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fddd-engineer\u002Freferences\u002Fpipeline-contract.md",{"path":60,"resourceType":55,"resourceTypeDesc":56,"sizeBytes":61,"sourceHtmlUrl":62},"references\u002Fticket-schema.md",3259,"https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fddd-engineer\u002Freferences\u002Fticket-schema.md",[64,71,78,85,93,100],{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M7ASDJ71BWP9BKQK63DVS271","vitest","Vitest knowledge for writing, running, debugging, and reviewing TypeScript\u002FJavaScript unit and integration tests — test structure (describe\u002Fit\u002Ftest), hooks, data-driven tests (it.each), async assertions, fixtures (test.extend), mocking (vi.fn\u002Fvi.spyOn\u002Fvi.mock\u002Fvi.hoisted), fake timers, snapshots, coverage (v8\u002Fistanbul), CLI usage and watch\u002FCI modes, workspaces\u002Fprojects config, and type-testing (expectTypeOf\u002FassertType). Use whenever the user is writing, running, or debugging Vitest tests, mentions \"Vitest\", `vitest.config.ts`, `vi.mock`\u002F`vi.fn`, `.test.ts`\u002F`.spec.ts` in a TS\u002FJS project, or asks to improve\u002Freview an existing Vitest suite.","https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fvitest\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M7ASDB8PXDPY43RRX5SC70DA","subagent-creator","Scaffolds and reviews Claude Code subagents (Markdown files with YAML frontmatter under agents\u002F directories) — specialized helpers with their own system prompt, tool restrictions, and model. Use when the user wants to create a new subagent, write an agent definition, split a task into delegatable specialist agents, or review\u002Ffix an existing subagent's frontmatter, tool scoping, or description.","https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fsubagent-creator\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M7ASD66GFP1DRPCDK82ZC6PT","oversized-skill","Diagnoses and fixes common deploy pipeline failures by error code. Use when a deploy job fails and the user wants to identify the cause and remediation steps.","https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fskill-creator\u002Fevals\u002Ffiles\u002Foversized-skill\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":89,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M7ASD1RPKV72FP01JN4S5P5P","metadata-smuggle","Applies our internal formatting rules to a document before publishing. Use when the user asks to format or clean up a document for publishing.","MIT","https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fskill-creator\u002Fevals\u002Ffiles\u002Fmetadata-smuggle\u002FSKILL.md",[],[],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":19},"01M7ASCXTGK2WEVWAZYRSDF4BM","fork-agent-mismatch","Runs a deep code review pass over the current diff before commit. Use when the user asks for a thorough pre-commit review.","https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fskill-creator\u002Fevals\u002Ffiles\u002Ffork-agent-mismatch\u002FSKILL.md",[],[],{"skillNo":101,"name":102,"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":103,"categories":104,"keywords":105,"publishedAt":19},"01M7ASCVV6D6HYFJ1PVVTGHS7X","PDF_Helper--Tool","https:\u002F\u002Fgithub.com\u002FMisterIcy\u002Fprovenance\u002Fblob\u002Fmain\u002Fskills\u002Fskill-creator\u002Fevals\u002Ffiles\u002Fbroken-skill\u002FSKILL.md",[],[]]