[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M78C6HAREGTXG2TAX0AYAXD4":3,"skill-same:01M78C6HAREGTXG2TAX0AYAXD4":101},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":15,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":13,"repositoryHtmlUrl":27,"repositoryHomepageUrl":28,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M78C6HAREGTXG2TAX0AYAXD4","cratis-screenplay-model-authoring",null,"Author, inspect, refactor and verify a Cratis Screenplay .play model through its typed AST and MCP workspace tools. Use for creating or changing model elements, navigating large split models, reviewing references\u002Fspecifications, revision-checked screen\u002FUI authoring proposals, or distinguishing source validity from executable\u002Fruntime readiness. Do not use for rendering a model into an application or hand-written Arc\u002FChronicle code.","Cratis\u002FPrologue","Cratis","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F16456213?v=4","01M6BXRH290K4C8RS4PTKK3HXQ","01M78C1YXTSPZGP4G1K2ZM6J1N",0,"MIT","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: cratis-screenplay-model-authoring\ndescription: Author, inspect, refactor and verify a Cratis Screenplay .play model through its typed AST and MCP workspace tools. Use for creating or changing model elements, navigating large split models, reviewing references\u002Fspecifications, revision-checked screen\u002FUI authoring proposals, or distinguishing source validity from executable\u002Fruntime readiness. Do not use for rendering a model into an application or hand-written Arc\u002FChronicle code.\nlicense: MIT\n---\n\u003C!-- cratis-ai-managed: skills\u002Fcratis-screenplay-model-authoring\u002FSKILL.md -->\n\n# Author a Screenplay model through its AST\n\nRead [preferred syntax](..\u002Fcratis-screenplay-toolchain\u002Freferences\u002Fpreferred-syntax.md) when writing or reviewing `.play`.\n\nPrefer the Screenplay MCP tools to text\u002Fregex edits: the compiler owns declaration,\nreference, hierarchy and source evidence for the declarative document set.\n\n**Valid source is not necessarily executable.** Stage owns rendering\u002Fruntime\nadmission. Read the [language reference](references\u002Flanguage-reference.md) for\nconstruct syntax, source-versus-execution states and a complete compiled model.\n\nFor screens, read `references\u002Fscreen-authoring-workflow.md`; use the\nsame model root or workspace export for all checks. Syntax: `cratis-screenplay-ui-composition`.\n\nRecord open modeling questions as `question` notes in committed\n`.screenplay\u002Fnotes.json`, not chat or `description` text. Never put secrets or\npersonal data in notes. Read [model notes](references\u002Fmodel-notes.md) when\nlisting, recording or resolving questions: `list-model-notes`,\n`propose-model-notes`, `read-proposal view=notes`, refs and orphan review.\nApplication, Module, Feature and Slice notes anchor before binding; members\nstill need binding and otherwise substitute their own slice. Inspect\n`containerIdentities` in `open-workspace` \u002F `workspace-state` for missing or\nambiguous identities; fix parse\u002Fplacement errors first. Read the notes reference\nfor refusals, `ambiguous: true`, typed targets and every `substitutions` entry.\n\n## Choosing a strategy\n\nRead [shop strategies and entropy](references\u002Fchoosing-generation.md) before\nediting concept generation or `generated identifier`.\n\n## When not to\n\nFor generation proofs\u002FPLAY0688 repairs, read [generation details](references\u002Fmcp-tools.md#concept-generation):\nuse `rules[]`, not parsed diagnostics (Requires Screenplay 4.173.0,\n`screenplay --version`; older servers omit proofs).\n\n## Locate and connect\n\nUse the declared `screenplay` MCP server (for example `open-workspace`).\n\nNever ask the user which folder to use before trying `open-workspace` without\narguments; do not invent a root. Read [MCP loop](references\u002Fmcp-loop.md) when connecting or switching roots;\nit owns fallback discovery, fixed-root limits and the initialization protocol.\nDiscover `tools\u002Flist` schemas.\n\nLook first for the project's existing `.play` files: their folder is the model.\nNew models go in `Source\u002F` or `src\u002F`, else root `Screenplay\u002F`, never `.cratis\u002F`\n(configuration\u002Fshared corpus only). The source is the single flow model;\nexplanatory Markdown belongs in repository documentation.\n\n`cratis ai install` manages `.cratis\u002Fai\u002F`, never model files. Select the\n`cratis\u002Fscreenplay` profile in the project's AI configuration. The corpus-owned\n`mcp-servers.json` declares the Screenplay server; supported client registration\nis owned by the Cratis CLI, which preserves other servers and user configuration.\n\nThe CLI-bundled MCP can be older than the standalone tool: discover schemas.\nRead [MCP loop](references\u002Fmcp-loop.md) for supported launch channels; do not\nsubstitute a global tool, Docker or copied managed configuration.\n\nIf MCP is unavailable, report it; regex edits are not compiler-validated AST\nediting. Existing standalone `screenplay mcp` users may retain that entry point.\n\n## Understand before editing\n\n1. Start with `describe-application` counts and paged logical children. A folder\n   of `.play` files is one application, not independent files.\n2. Use `search-declarations`, `declaration-details`, `find-references` and\n   `dependencies` to locate the owning declaration and affected references.\n   Before a property change, read [property links](references\u002Fproperty-links.md):\n   `kind: \"Property\"`, `Owner.member` \u002F `Owner.generationN.member`, direct mappings.\n   `complete: false` is incomplete evidence, never a successful empty list.\n   Edit through MCP proposals, not hand-edited text; [generation](references\u002Fmcp-tools.md#concept-generation).\n3. Read diagnostics and coverage limitations. Identify subjects by code and typed\n   `arguments`, never by parsing message text. Read [diagnostic arguments](..\u002Fcratis-screenplay-toolchain\u002Freferences\u002Fdiagnostic-arguments.md)\n   for kinds, PLAY0524 step paths\u002Findices, unknown fields and older responses. Use `dependency-graph` for container\n   edges, declared-dependency coverage, cycles and advisory order; the board's\n   dependency-map toggle exposes this evidence visually. Unknown or ambiguous\n   bindings are not an empty, successful dependency graph.\n4. Use `find-fixtures` for value occurrences (including example\u002Foverride origins\n   and event routes) and `find-assertion-gaps` for authored assertion gaps. Neither\n   executes specifications; use `run-specifications` for that. `diagnostics` accepts `scope` and opt-in `checks`;\n   skipped completeness checks are not passes. Details: `references\u002Fmcp-tools.md`.\n5. Echo `sourceRevision` on subsequent pages. Restart the query after drift;\n   never combine pages from different snapshots.\n6. For screens, also discover the template, form, toolbar, package\u002Ficon and\n   outlet declarations that provide context. A component binding with no traced\n   data source is not a shortcut; fix the model source.\n\nScope reads to the relevant container\u002Fdocument, not the whole merged AST.\nSize-limit refusals require narrower reads, not truncation.\n\n## Application modes and described handler intent\n\nStart new models with `reaction batches atomic`; preserve it through proposals.\nRead [language admission](..\u002Fcratis-screenplay-toolchain\u002Freferences\u002Flanguage-admission.md)\nfor atomic\u002Fprotected modes and Stage refusals. Read [intent inventories](references\u002Fmcp-tools.md#dependencies-and-intent-inventories)\nfor handler views: `executableReady` means binding, not an implemented body.\n\n## Compare models\n\nCompare with `main` through `semantic-diff` worktree `path`, not two exports.\nRead [comparison sources](references\u002Fmcp-tools.md#semantic-revision-comparison) for paging.\n\n## Make one coherent proposal\n\n**One edit strategy.** Prefer discovered identity-preserving operations\n(`propose-rename`, `propose-repair`, `propose-extract-inline-event`, `propose-ast`).\nBounded text edits must leave catalog addresses unchanged (descriptions, rule\nbodies, existing-member mappings), never a way around an MCP refusal. When `.screenplay\u002Fidentities.json`\nexists, adding, removing or renaming a declaration, command or read-model property, query,\nquery argument or specification goes through MCP, and an `id` pin is not enough for a rename or move. A request naming\na rename, move or removal is the approval: do not ask again at `apply`. Without MCP, text-rename\nwith `id \"\u003COld>\"` pins and an identity note; only with `identities.json` and no MCP, return the\nrequest to the owning session, saying first it is NOT done.\nProposals are connection-local (16 cap); only the owner proposes\u002Fapplies.\nOthers send an edit request ([MCP loop](references\u002Fmcp-loop.md)).\n\n- Apply is not crash-atomic across files: it journals its inverse first. The model\n  root must be trusted and exclusively writable during effects.\n\n1. `open-workspace` obtains workspace\u002Fcatalog revisions and restores root-local\n   identity state. `read-workspace` locates document identities.\n2. `read-ast` returns original occurrence handles and existing semantic IDs.\n   Query `syntax-schema` for exact node members before constructing typed input.\n3. Prefer `propose-rename` and `propose-move`: they carry identities and notes\n   even on unbound models. Manual `propose-ast` \u002F `propose-source` renames need\n   `semanticRenames`; removal needs retirement and reports `orphanedNotes`.\n   Never substitute global string replacement for an explicit refusal.\n4. Prefer `propose-property` for property add\u002Frename\u002Fremove, not hand-composed\n   `propose-ast` batches. Inspect the chain and never guess past a refusal:\n   [property links](references\u002Fproperty-links.md). Otherwise use `propose-source`\n   (whole text) or `propose-ast` (typed edits); `propose` is executable-only.\n   Default hand-written source to `PreserveExactSource` (Requires Screenplay 4.158.0,\n   `screenplay --version`); read `references\u002Fmcp-loop.md` when choosing formatting.\n5. Keep reference policy `Safe`. Use `Draft` only for deliberately requested\n   unresolved model debt and report that debt; it does not waive structural,\n   identity or binding-protection checks. Batch independent reference removals\u002Fadditions;\n   unmatched same-text pairs refuse as ambiguous. Preserve moves with `propose-move`\n   or semantic migrations. String `conflicts[].document`\u002F`path` locate refusals:\n   [reference continuity and version gate](references\u002Fmcp-loop.md#reference-continuity-and-located-conflicts).\n6. Report `summary`, inspect severity errors\u002Fcodes rather than counts\n   (Requires Screenplay 4.156.0, `screenplay --version`); read `references\u002Fmcp-loop.md`\n   when reviewing PLAY0288, comment loss or introduced executable errors. Inspect `read-proposal` before\u002Fafter bytes and diagnostics, including the\n   identity-state change through `workspace-state`. Review `orphanedNotes` on\n   source proposals; reanchor or resolve affected questions before applying\n   ([notes and orphan impact](references\u002Fmodel-notes.md#source-proposals-can-orphan-notes)).\n7. Run `run-specifications` with `proposalId` before applying; record V4 separately.\n   PLAY0285 unbinds the whole model. Read the [MCP tool guide](references\u002Fmcp-tools.md#run-reference-specifications)\n   for candidate paging and the required version.\n8. `apply` only the reviewed server-produced proposal within the user's requested\n   change. The proposal ID and before revisions are required; stale state rejects.\n   Apply after checks if requested; otherwise show the proposal and ask once.\n9. When the host draws views, `visualize-model` shows the board. It follows the files\n   on disk while open, so after an apply do not re-request it; just say what changed.\n\nHandles are revision-bound, not durable IDs. Modules\u002Ffeatures span fragments;\nafter apply fetch fresh handles\u002Frevisions.\n\nOn refusal, use `failureKind`, never message text ([failures\u002Flimits](references\u002Fmcp-loop.md#failures-use-failurekind-never-message-text)).\nAfter revision\u002Fdisk\u002Fidentity drift, reopen and re-propose. For `ApplyOutcomeUnknown`\nor EOF during `apply`, **never retry apply**: reconnect, inspect `workspace-state`\nand ask. Never delete `.screenplay\u002Fpending.json`. Limits reject, never truncate.\n\nChoose formatting and disclose dropped comments before apply:\n[formatting\u002Flimits](references\u002Fmcp-loop.md#formatting-and-limits) covers\n`PreserveTrivia`, canonical consent and byte preservation.\n\n`identityMigrationIssues`: read the [MCP loop](references\u002Fmcp-loop.md).\n\n## Repair or extract before rewriting\n\nRead `read-workspace` with `view: \"repairs\"` at the same revision as diagnostics and pass the repair to `propose-repair`; use `propose-extract-inline-event` before adding generations, or let a `remove` with `newGeneration: true` extract the inline event itself, as [property links](references\u002Fproperty-links.md) describes. `PLAY0469` changes the event contract, so ask about stored contracts first. No automatic generation-evolution repair ships. `PLAY0479`: migrate optionality spelling with `PreserveTrivia`. `eventNeverPersisted: true` cannot unfreeze admitted revisions. Differing event pins carry `storedName` in ESM v16, shared by all generations and unique (PLAY0472). Read [references\u002Frepairs.md](references\u002Frepairs.md) when fixing `PLAY0166`, `PLAY0478`, `PLAY0516` or `PLAY0683`, or admitting a [release baseline](references\u002Frelease-baseline.md). Read [references\u002Fdestination-intent-repair.md](references\u002Fdestination-intent-repair.md) when `PLAY0470` needs a repair.\n\n## Choose a readable layout\n\nUse `recommend-layout` for options and review an `expand-layout` proposal; layout is not application semantics. Use one file per module when modules are simple. Use one file per slice when a large model needs isolated diffs. Each level above the slices is a barrel file. Read [references\u002Flayout.md](references\u002Flayout.md) when splitting files or writing import barrels.\n\n## Preserve state and recover explicitly\n\nFix binding errors through MCP proposals, not outside MCP: reopening after an\nexternal fix can require reconciliation of a persisted partial catalog.\nCommit `.screenplay\u002Fidentities.json` with the model; only `apply` writes it. Never delete or overwrite identity state, mint IDs on restart, or text-rename identity-affecting declarations. A pending journal blocks normal work: inspect `workspace-state` and use `recover-workspace` only for the identified interrupted operation. Only `apply` \u002F `recover-workspace` write source\u002Fcatalog. `readiness.authoringAccepted` says the source is valid Screenplay; `executableReady` is separate. Read [references\u002Fstate-and-recovery.md](references\u002Fstate-and-recovery.md) for root conflicts, journals and readiness.\n\n## Code attachments\n\nScreenplay never compiles\u002Fruns opaque code attachments or `file` constraints;\na screen's `file` realizes UI instead. Read [MCP code attachments](references\u002Fmcp-tools.md#code-attachments)\nfor paths, PLAY0430–PLAY0434 and implementation-requirements views; hashes\u002Fmaps\nare not behavioral evidence.\n\n## Verify and report honestly\n\n- Require zero source errors; investigate warnings, reference safety, identity\n  continuity and exact intended changes.\n- Report which of the four states you checked: **parsed**, **bound**,\n  **reference-executed**, **target-executed**. When distinguishing source validity from execution, read the\n  [language reference](references\u002Flanguage-reference.md#source-validity-is-not-execution);\n  it defines them and maps them onto V1 to V5. `screenplay --warnaserror` checks\n  syntax plus the consistency rules (`PLAY0282`-`PLAY0294`) but never binds, so a\n  clean run is not a bound or executable result.\n- Never drop or stub source contracts to bind; the bundled compiler and renderer can be older than the standalone tool. Read [references\u002Flanguage-reference.md](references\u002Flanguage-reference.md) for the ESM version at which each construct binds.\n\n- Unsupported capabilities never justify dropping constructs or inventing stubs.\n  Validate intended execution with the owning downstream runtime.\n- Model rejection cases as specifications; assertion presence is not execution.\n- Check installed `tools\u002Flist`\u002Fschemas rather than assuming a remembered tool count.\n\nThe ordinary CLI remains useful for whole-folder validation:\n\n```shell\ncratis screenplay validate \u003Cmodel-folder> --warnings-as-errors\n```\n\nKeep screen transcripts in `.ai-work\u002F`; check the same source, never an app that\nsilently drops authored UI. Use the Screenplay-owned canonical\ncorpus fixture path instead of copying fixtures here.\n\nNever invent syntax or capabilities.\n\n## Route near misses\n\n| Need | Skill |\n| --- | --- |\n| Discovering the domain and event-modeling method | `cratis-screenplay-event-modeling` |\n| Commands, events, validation and concurrency | `cratis-screenplay-command-surface` |\n| Projections and reducers | `cratis-screenplay-projections` |\n| Read models, queries and screens | `cratis-screenplay-read-surface` |\n| UI composition, bindings, navigation and themes | `cratis-screenplay-ui-composition` |\n| Captures, reactions and triggers | `cratis-screenplay-captures-and-reactions` |\n| Behavioral examples and assertions | `cratis-screenplay-specifications` |\n| Rendering\u002Frunning an admitted model | `cratis-stage-rendering-and-sandbox` |\n| Lifecycle and model-first decision | `cratis-screenplay-modeling-lifecycle` |\n| Compiler, CLI, MCP and diagnostics | `cratis-screenplay-toolchain` |\n| Event-model diagrams rather than .play source | `cratis-event-model-diagram` |\n","\u003C!-- cratis-ai-managed: skills\u002Fcratis-screenplay-model-authoring\u002FSKILL.md -->\n\n# Author a Screenplay model through its AST\n\nRead [preferred syntax](..\u002Fcratis-screenplay-toolchain\u002Freferences\u002Fpreferred-syntax.md) when writing or reviewing `.play`.\n\nPrefer the Screenplay MCP tools to text\u002Fregex edits: the compiler owns declaration,\nreference, hierarchy and source evidence for the declarative document set.\n\n**Valid source is not necessarily executable.** Stage owns rendering\u002Fruntime\nadmission. Read the [language reference](references\u002Flanguage-reference.md) for\nconstruct syntax, source-versus-execution states and a complete compiled model.\n\nFor screens, read `references\u002Fscreen-authoring-workflow.md`; use the\nsame model root or workspace export for all checks. Syntax: `cratis-screenplay-ui-composition`.\n\nRecord open modeling questions as `question` notes in committed\n`.screenplay\u002Fnotes.json`, not chat or `description` text. Never put secrets or\npersonal data in notes. Read [model notes](references\u002Fmodel-notes.md) when\nlisting, recording or resolving questions: `list-model-notes`,\n`propose-model-notes`, `read-proposal view=notes`, refs and orphan review.\nApplication, Module, Feature and Slice notes anchor before binding; members\nstill need binding and otherwise substitute their own slice. Inspect\n`containerIdentities` in `open-workspace` \u002F `workspace-state` for missing or\nambiguous identities; fix parse\u002Fplacement errors first. Read the notes reference\nfor refusals, `ambiguous: true`, typed targets and every `substitutions` entry.\n\n## Choosing a strategy\n\nRead [shop strategies and entropy](references\u002Fchoosing-generation.md) before\nediting concept generation or `generated identifier`.\n\n## When not to\n\nFor generation proofs\u002FPLAY0688 repairs, read [generation details](references\u002Fmcp-tools.md#concept-generation):\nuse `rules[]`, not parsed diagnostics (Requires Screenplay 4.173.0,\n`screenplay --version`; older servers omit proofs).\n\n## Locate and connect\n\nUse the declared `screenplay` MCP server (for example `open-workspace`).\n\nNever ask the user which folder to use before trying `open-workspace` without\narguments; do not invent a root. Read [MCP loop](references\u002Fmcp-loop.md) when connecting or switching roots;\nit owns fallback discovery, fixed-root limits and the initialization protocol.\nDiscover `tools\u002Flist` schemas.\n\nLook first for the project's existing `.play` files: their folder is the model.\nNew models go in `Source\u002F` or `src\u002F`, else root `Screenplay\u002F`, never `.cratis\u002F`\n(configuration\u002Fshared corpus only). The source is the single flow model;\nexplanatory Markdown belongs in repository documentation.\n\n`cratis ai install` manages `.cratis\u002Fai\u002F`, never model files. Select the\n`cratis\u002Fscreenplay` profile in the project's AI configuration. The corpus-owned\n`mcp-servers.json` declares the Screenplay server; supported client registration\nis owned by the Cratis CLI, which preserves other servers and user configuration.\n\nThe CLI-bundled MCP can be older than the standalone tool: discover schemas.\nRead [MCP loop](references\u002Fmcp-loop.md) for supported launch channels; do not\nsubstitute a global tool, Docker or copied managed configuration.\n\nIf MCP is unavailable, report it; regex edits are not compiler-validated AST\nediting. Existing standalone `screenplay mcp` users may retain that entry point.\n\n## Understand before editing\n\n1. Start with `describe-application` counts and paged logical children. A folder\n   of `.play` files is one application, not independent files.\n2. Use `search-declarations`, `declaration-details`, `find-references` and\n   `dependencies` to locate the owning declaration and affected references.\n   Before a property change, read [property links](references\u002Fproperty-links.md):\n   `kind: \"Property\"`, `Owner.member` \u002F `Owner.generationN.member`, direct mappings.\n   `complete: false` is incomplete evidence, never a successful empty list.\n   Edit through MCP proposals, not hand-edited text; [generation](references\u002Fmcp-tools.md#concept-generation).\n3. Read diagnostics and coverage limitations. Identify subjects by code and typed\n   `arguments`, never by parsing message text. Read [diagnostic arguments](..\u002Fcratis-screenplay-toolchain\u002Freferences\u002Fdiagnostic-arguments.md)\n   for kinds, PLAY0524 step paths\u002Findices, unknown fields and older responses. Use `dependency-graph` for container\n   edges, declared-dependency coverage, cycles and advisory order; the board's\n   dependency-map toggle exposes this evidence visually. Unknown or ambiguous\n   bindings are not an empty, successful dependency graph.\n4. Use `find-fixtures` for value occurrences (including example\u002Foverride origins\n   and event routes) and `find-assertion-gaps` for authored assertion gaps. Neither\n   executes specifications; use `run-specifications` for that. `diagnostics` accepts `scope` and opt-in `checks`;\n   skipped completeness checks are not passes. Details: `references\u002Fmcp-tools.md`.\n5. Echo `sourceRevision` on subsequent pages. Restart the query after drift;\n   never combine pages from different snapshots.\n6. For screens, also discover the template, form, toolbar, package\u002Ficon and\n   outlet declarations that provide context. A component binding with no traced\n   data source is not a shortcut; fix the model source.\n\nScope reads to the relevant container\u002Fdocument, not the whole merged AST.\nSize-limit refusals require narrower reads, not truncation.\n\n## Application modes and described handler intent\n\nStart new models with `reaction batches atomic`; preserve it through proposals.\nRead [language admission](..\u002Fcratis-screenplay-toolchain\u002Freferences\u002Flanguage-admission.md)\nfor atomic\u002Fprotected modes and Stage refusals. Read [intent inventories](references\u002Fmcp-tools.md#dependencies-and-intent-inventories)\nfor handler views: `executableReady` means binding, not an implemented body.\n\n## Compare models\n\nCompare with `main` through `semantic-diff` worktree `path`, not two exports.\nRead [comparison sources](references\u002Fmcp-tools.md#semantic-revision-comparison) for paging.\n\n## Make one coherent proposal\n\n**One edit strategy.** Prefer discovered identity-preserving operations\n(`propose-rename`, `propose-repair`, `propose-extract-inline-event`, `propose-ast`).\nBounded text edits must leave catalog addresses unchanged (descriptions, rule\nbodies, existing-member mappings), never a way around an MCP refusal. When `.screenplay\u002Fidentities.json`\nexists, adding, removing or renaming a declaration, command or read-model property, query,\nquery argument or specification goes through MCP, and an `id` pin is not enough for a rename or move. A request naming\na rename, move or removal is the approval: do not ask again at `apply`. Without MCP, text-rename\nwith `id \"\u003COld>\"` pins and an identity note; only with `identities.json` and no MCP, return the\nrequest to the owning session, saying first it is NOT done.\nProposals are connection-local (16 cap); only the owner proposes\u002Fapplies.\nOthers send an edit request ([MCP loop](references\u002Fmcp-loop.md)).\n\n- Apply is not crash-atomic across files: it journals its inverse first. The model\n  root must be trusted and exclusively writable during effects.\n\n1. `open-workspace` obtains workspace\u002Fcatalog revisions and restores root-local\n   identity state. `read-workspace` locates document identities.\n2. `read-ast` returns original occurrence handles and existing semantic IDs.\n   Query `syntax-schema` for exact node members before constructing typed input.\n3. Prefer `propose-rename` and `propose-move`: they carry identities and notes\n   even on unbound models. Manual `propose-ast` \u002F `propose-source` renames need\n   `semanticRenames`; removal needs retirement and reports `orphanedNotes`.\n   Never substitute global string replacement for an explicit refusal.\n4. Prefer `propose-property` for property add\u002Frename\u002Fremove, not hand-composed\n   `propose-ast` batches. Inspect the chain and never guess past a refusal:\n   [property links](references\u002Fproperty-links.md). Otherwise use `propose-source`\n   (whole text) or `propose-ast` (typed edits); `propose` is executable-only.\n   Default hand-written source to `PreserveExactSource` (Requires Screenplay 4.158.0,\n   `screenplay --version`); read `references\u002Fmcp-loop.md` when choosing formatting.\n5. Keep reference policy `Safe`. Use `Draft` only for deliberately requested\n   unresolved model debt and report that debt; it does not waive structural,\n   identity or binding-protection checks. Batch independent reference removals\u002Fadditions;\n   unmatched same-text pairs refuse as ambiguous. Preserve moves with `propose-move`\n   or semantic migrations. String `conflicts[].document`\u002F`path` locate refusals:\n   [reference continuity and version gate](references\u002Fmcp-loop.md#reference-continuity-and-located-conflicts).\n6. Report `summary`, inspect severity errors\u002Fcodes rather than counts\n   (Requires Screenplay 4.156.0, `screenplay --version`); read `references\u002Fmcp-loop.md`\n   when reviewing PLAY0288, comment loss or introduced executable errors. Inspect `read-proposal` before\u002Fafter bytes and diagnostics, including the\n   identity-state change through `workspace-state`. Review `orphanedNotes` on\n   source proposals; reanchor or resolve affected questions before applying\n   ([notes and orphan impact](references\u002Fmodel-notes.md#source-proposals-can-orphan-notes)).\n7. Run `run-specifications` with `proposalId` before applying; record V4 separately.\n   PLAY0285 unbinds the whole model. Read the [MCP tool guide](references\u002Fmcp-tools.md#run-reference-specifications)\n   for candidate paging and the required version.\n8. `apply` only the reviewed server-produced proposal within the user's requested\n   change. The proposal ID and before revisions are required; stale state rejects.\n   Apply after checks if requested; otherwise show the proposal and ask once.\n9. When the host draws views, `visualize-model` shows the board. It follows the files\n   on disk while open, so after an apply do not re-request it; just say what changed.\n\nHandles are revision-bound, not durable IDs. Modules\u002Ffeatures span fragments;\nafter apply fetch fresh handles\u002Frevisions.\n\nOn refusal, use `failureKind`, never message text ([failures\u002Flimits](references\u002Fmcp-loop.md#failures-use-failurekind-never-message-text)).\nAfter revision\u002Fdisk\u002Fidentity drift, reopen and re-propose. For `ApplyOutcomeUnknown`\nor EOF during `apply`, **never retry apply**: reconnect, inspect `workspace-state`\nand ask. Never delete `.screenplay\u002Fpending.json`. Limits reject, never truncate.\n\nChoose formatting and disclose dropped comments before apply:\n[formatting\u002Flimits](references\u002Fmcp-loop.md#formatting-and-limits) covers\n`PreserveTrivia`, canonical consent and byte preservation.\n\n`identityMigrationIssues`: read the [MCP loop](references\u002Fmcp-loop.md).\n\n## Repair or extract before rewriting\n\nRead `read-workspace` with `view: \"repairs\"` at the same revision as diagnostics and pass the repair to `propose-repair`; use `propose-extract-inline-event` before adding generations, or let a `remove` with `newGeneration: true` extract the inline event itself, as [property links](references\u002Fproperty-links.md) describes. `PLAY0469` changes the event contract, so ask about stored contracts first. No automatic generation-evolution repair ships. `PLAY0479`: migrate optionality spelling with `PreserveTrivia`. `eventNeverPersisted: true` cannot unfreeze admitted revisions. Differing event pins carry `storedName` in ESM v16, shared by all generations and unique (PLAY0472). Read [references\u002Frepairs.md](references\u002Frepairs.md) when fixing `PLAY0166`, `PLAY0478`, `PLAY0516` or `PLAY0683`, or admitting a [release baseline](references\u002Frelease-baseline.md). Read [references\u002Fdestination-intent-repair.md](references\u002Fdestination-intent-repair.md) when `PLAY0470` needs a repair.\n\n## Choose a readable layout\n\nUse `recommend-layout` for options and review an `expand-layout` proposal; layout is not application semantics. Use one file per module when modules are simple. Use one file per slice when a large model needs isolated diffs. Each level above the slices is a barrel file. Read [references\u002Flayout.md](references\u002Flayout.md) when splitting files or writing import barrels.\n\n## Preserve state and recover explicitly\n\nFix binding errors through MCP proposals, not outside MCP: reopening after an\nexternal fix can require reconciliation of a persisted partial catalog.\nCommit `.screenplay\u002Fidentities.json` with the model; only `apply` writes it. Never delete or overwrite identity state, mint IDs on restart, or text-rename identity-affecting declarations. A pending journal blocks normal work: inspect `workspace-state` and use `recover-workspace` only for the identified interrupted operation. Only `apply` \u002F `recover-workspace` write source\u002Fcatalog. `readiness.authoringAccepted` says the source is valid Screenplay; `executableReady` is separate. Read [references\u002Fstate-and-recovery.md](references\u002Fstate-and-recovery.md) for root conflicts, journals and readiness.\n\n## Code attachments\n\nScreenplay never compiles\u002Fruns opaque code attachments or `file` constraints;\na screen's `file` realizes UI instead. Read [MCP code attachments](references\u002Fmcp-tools.md#code-attachments)\nfor paths, PLAY0430–PLAY0434 and implementation-requirements views; hashes\u002Fmaps\nare not behavioral evidence.\n\n## Verify and report honestly\n\n- Require zero source errors; investigate warnings, reference safety, identity\n  continuity and exact intended changes.\n- Report which of the four states you checked: **parsed**, **bound**,\n  **reference-executed**, **target-executed**. When distinguishing source validity from execution, read the\n  [language reference](references\u002Flanguage-reference.md#source-validity-is-not-execution);\n  it defines them and maps them onto V1 to V5. `screenplay --warnaserror` checks\n  syntax plus the consistency rules (`PLAY0282`-`PLAY0294`) but never binds, so a\n  clean run is not a bound or executable result.\n- Never drop or stub source contracts to bind; the bundled compiler and renderer can be older than the standalone tool. Read [references\u002Flanguage-reference.md](references\u002Flanguage-reference.md) for the ESM version at which each construct binds.\n\n- Unsupported capabilities never justify dropping constructs or inventing stubs.\n  Validate intended execution with the owning downstream runtime.\n- Model rejection cases as specifications; assertion presence is not execution.\n- Check installed `tools\u002Flist`\u002Fschemas rather than assuming a remembered tool count.\n\nThe ordinary CLI remains useful for whole-folder validation:\n\n```shell\ncratis screenplay validate \u003Cmodel-folder> --warnings-as-errors\n```\n\nKeep screen transcripts in `.ai-work\u002F`; check the same source, never an app that\nsilently drops authored UI. Use the Screenplay-owned canonical\ncorpus fixture path instead of copying fixtures here.\n\nNever invent syntax or capabilities.\n\n## Route near misses\n\n| Need | Skill |\n| --- | --- |\n| Discovering the domain and event-modeling method | `cratis-screenplay-event-modeling` |\n| Commands, events, validation and concurrency | `cratis-screenplay-command-surface` |\n| Projections and reducers | `cratis-screenplay-projections` |\n| Read models, queries and screens | `cratis-screenplay-read-surface` |\n| UI composition, bindings, navigation and themes | `cratis-screenplay-ui-composition` |\n| Captures, reactions and triggers | `cratis-screenplay-captures-and-reactions` |\n| Behavioral examples and assertions | `cratis-screenplay-specifications` |\n| Rendering\u002Frunning an admitted model | `cratis-stage-rendering-and-sandbox` |\n| Lifecycle and model-first decision | `cratis-screenplay-modeling-lifecycle` |\n| Compiler, CLI, MCP and diagnostics | `cratis-screenplay-toolchain` |\n| Event-model diagrams rather than .play source | `cratis-event-model-diagram` |\n",{"name":5,"license":14,"description":7},"main","feed400071bf3f9f4410a956d73413daa6e1b334",".cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue","https:\u002F\u002Fwww.cratis.io\u002Fprologue\u002F","https:\u002F\u002Fgithub.com\u002FCratis","https:\u002F\u002Fwww.cratis.io","Open-source (MIT) event sourcing and CQRS for .NET and beyond: Chronicle, Arc, Components and Screenplay.","2026-11-14 06:37:10",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add Cratis\u002FPrologue --skill cratis-screenplay-model-authoring",{"tool":39,"command":40},"bunx","bunx skills add Cratis\u002FPrologue --skill cratis-screenplay-model-authoring",{"tool":42,"command":43},"pnpm","pnpm dlx skills add Cratis\u002FPrologue --skill cratis-screenplay-model-authoring","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002FSKILL.md",true,[47,51,57,61,65,69,73,77,81,85,89,93,97],{"path":48,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":15},"LICENSE","许可证",212,{"path":52,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fchoosing-generation.md","REFERENCE","参考文档",2744,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Fchoosing-generation.md",{"path":58,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fdestination-intent-repair.md",5644,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Fdestination-intent-repair.md",{"path":62,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Flanguage-reference.md",28705,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Flanguage-reference.md",{"path":66,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":67,"sourceHtmlUrl":68},"references\u002Flayout.md",1672,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Flayout.md",{"path":70,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":71,"sourceHtmlUrl":72},"references\u002Fmcp-loop.md",26554,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Fmcp-loop.md",{"path":74,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fmcp-tools.md",57015,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Fmcp-tools.md",{"path":78,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":79,"sourceHtmlUrl":80},"references\u002Fmodel-notes.md",14769,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Fmodel-notes.md",{"path":82,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":83,"sourceHtmlUrl":84},"references\u002Fproperty-links.md",13293,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Fproperty-links.md",{"path":86,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":87,"sourceHtmlUrl":88},"references\u002Frelease-baseline.md",8653,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Frelease-baseline.md",{"path":90,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":91,"sourceHtmlUrl":92},"references\u002Frepairs.md",3293,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Frepairs.md",{"path":94,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":95,"sourceHtmlUrl":96},"references\u002Fscreen-authoring-workflow.md",15858,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Fscreen-authoring-workflow.md",{"path":98,"resourceType":53,"resourceTypeDesc":54,"sizeBytes":99,"sourceHtmlUrl":100},"references\u002Fstate-and-recovery.md",3338,"https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fcratis-screenplay-model-authoring\u002Freferences\u002Fstate-and-recovery.md",[102,110,118,126,134,142],{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":106,"sourceHtmlUrl":107,"categories":108,"keywords":109,"publishedAt":19},"01M78CCRM9594FCYFMCY5W137G","vectorization","Designs, implements, optimizes, and reviews SIMD code in .NET: vectorizing scalar loops with TensorPrimitives, Vector64, Vector128, Vector256, Vector512, or hardware intrinsics, and reviewing existing SIMD code, including the generic Vector type, for contract equivalence, tail handling, memory safety, portability, fallbacks, and measured performance. Use when a .NET loop or SIMD routine is being written or reviewed. Not for performance work unrelated to SIMD.","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fvectorization\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fvectorization\u002FSKILL.md",[],[],{"skillNo":111,"name":112,"nameZh":6,"description":113,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":114,"sourceHtmlUrl":115,"categories":116,"keywords":117,"publishedAt":19},"01M78CCJKGDNSDNCMA92PV8GCR","target-authoring","Canonical patterns for writing custom MSBuild targets: DependsOnTargets versus BeforeTargets and AfterTargets, the Build-to-CoreBuild three-level pattern, and the XxxDependsOn chain-extension pattern. Use when diagnosing or fixing custom-target anti-patterns: Directory.Build.targets redefining SDK targets, targets that replace CompileDependsOn instead of extending it, Outputs versus Returns misuse that returns stale results, missing Inputs\u002FOutputs causing needless rebuilds, or missing FileWrites registration. Not for incremental build tuning or general anti-pattern review (use the msbuild skill), or non-MSBuild build systems.","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Ftarget-authoring\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Ftarget-authoring\u002FSKILL.md",[],[],{"skillNo":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":122,"sourceHtmlUrl":123,"categories":124,"keywords":125,"publishedAt":19},"01M78CCCWS16JZA36AGB5YVMAN","run-tests","Chooses and runs the repository-compatible .NET test command: classic, VSTest, bridged MTP, or native MTP syntax for MSTest, xUnit, NUnit, and TUnit. Use before running .NET tests or answering with a test command or flags: run the tests, one test, class, category, trait or target framework, combined filters, --filter-query, --no-build, --diag, TRX, coverage collection, crash or hang dumps, filter errors, or unrecognized options. Not for platform identification alone (use platform-detection), writing or debugging test code, interpreting a coverage report (use coverage-analysis), CI investigation, migration, or a persistent hot-reload loop.","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Frun-tests\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Frun-tests\u002FSKILL.md",[],[],{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":130,"sourceHtmlUrl":131,"categories":132,"keywords":133,"publishedAt":19},"01M78CC6VC08Y3WTN4T784DYDY","profiling","Uses the free official .NET diagnostics CLI tools for profiling and runtime investigation in a repository: CPU, GC pressure, allocation growth, exception storms, lock contention, and slow code. Use when a .NET application needs performance or runtime profiling. Not for replacing realistic performance or load tests with ad-hoc tracing, production heap collection without a pause-risk plan, crash dumps (use dump-collect), or benchmark design (use microbenchmarking).","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fprofiling\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fprofiling\u002FSKILL.md",[],[],{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":138,"sourceHtmlUrl":139,"categories":140,"keywords":141,"publishedAt":19},"01M78CC14WNAP1Y77ZMM4VD29A","platform-detection","Identifies a .NET project's test platform, framework, command mode, and SDK-style versus classic project system, resolving global.json, project, packages.config, Directory.Build.props, and Directory.Packages.props precedence for MSTest, xUnit, NUnit, and TUnit. Use when asking \"which test platform or framework?\", \"VSTest or MTP?\", or \"what runner does this project use?\", including bridge settings, UseVSTest opt-outs, and conflicting VSTest\u002FMTP configuration. Not for running or filtering tests, commands, flags, TRX, dumps, or test-command errors (use run-tests), or for hot-reload or migration requests.","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fplatform-detection\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Fplatform-detection\u002FSKILL.md",[],[],{"skillNo":143,"name":144,"nameZh":6,"description":145,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":146,"sourceHtmlUrl":147,"categories":148,"keywords":149,"publishedAt":19},"01M78CBVQKQZ6C9VZAD48GB8AC","orleans","Designs Microsoft Orleans systems from each primitive's purpose and failure model: grains, digital twins, state versus databases, transactions, messaging, streams, timers, reminders, Durable Jobs, stateless workers, grain services, startup, and hosting. Use when a task needs an Orleans design decision: inspect version and topology, choose the primitive, implement, and validate. Not for other actor stacks, batch processing, relational-only CRUD, or advice without an Orleans decision.","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Forleans\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FCratis\u002FPrologue\u002Fblob\u002Fmain\u002F.cratis\u002Fai\u002Fskills\u002Forleans\u002FSKILL.md",[],[]]