[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C44H2QWPA9N0TJZ2302XHT":3,"skill-same:01M6C44H2QWPA9N0TJZ2302XHT":61},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":29,"linkOnly":30,"markdownContent":31,"markdownBody":32,"frontMatter":33,"refName":34,"commitSha":35,"skillDirectory":36,"forksCount":37,"repositoryHtmlUrl":38,"repositoryHomepageUrl":6,"ownerHtmlUrl":39,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":40,"install":41,"resources":54},"01M6C44H2QWPA9N0TJZ2302XHT","engagement-record-emission",null,"Marker contract for Segment-A maintainer-evidence and cross-session engagement-state preservation. Use when an agent exits its phase. DO NOT USE FOR: runtime code execution, test writing, or PR creation.","Grimblaz\u002Fagent-orchestra","Grimblaz","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F74755849?v=4","01M6C430M9XZDHGHPWQVTQ3KMA","01M6C430RJY3MMX0QAZ3X8DR3Y",3,"MIT","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fengagement-record-emission\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28],"engagement","engagement record emission","writing","test","pr","code","state","execution","session","use","evidence","2026-11-10 10:12:30",false,"---\nname: engagement-record-emission\ndescription: \"Marker contract for Segment-A maintainer-evidence and cross-session engagement-state preservation. Use when an agent exits its phase. DO NOT USE FOR: runtime code execution, test writing, or PR creation.\"\n---\n\n# Engagement Record Emission\n\nThis skill defines the `SMC-20` engagement-record marker contract and schema used to preserve agentic engagement state across sessions.\n\n## When to Use\n\nUse this skill when an upstream agent (Experience-Owner, Solution-Designer, or Issue-Planner) exits its respective phase. It provides a structured, durable way to record the decisions, classifications, and engineer choices made during the phase, allowing subsequent sessions or downstream agents to skip repetitive questioning on settled decisions.\n\n**Also a writer (#974): the open-for-work conversation.** `skills\u002Fopen-for-work\u002FSKILL.md` records its checkpoint decisions — the worth-it doors, the affirmation gate, and brief approval on the routine arm — into `\u003C!-- engagement-record-experience-{ID} -->` at `schema_version: 2`, because that conversation *is* the experience-replacement for standalone work (#957 D5). It is not an upstream agent, so it is named here explicitly rather than inferred, on the same footing as the `orchestration`-phase carve-out below. **Cumulative-write obligation**: that conversation reaches three checkpoints at different moments, and this family is `post-new` with latest-createdAt-wins on read — so a later marker carrying only the later decision orphans the earlier one. Write one marker at conversation exit carrying every decision, or make each later write carry the full accumulated set.\n\n## When Not to Use\n\nDo not use this skill for runtime code execution, test writing, or PR creation. It is strictly a cognitive-evidence and state-preservation contract. Do not emit these markers in downstream phases. **Exception (v2 carve-out, #655 S3)**: the `review` phase is explicitly permitted to emit `\u003C!-- engagement-record-review-{PR} -->` as a PR comment after the review judge verdict; this marker is PR-keyed (not issue-keyed) and governed by SMC-23.\n\n## Marker Shape\n\nEngagement records are emitted as GitHub issue comments containing a YAML code block wrapped inside a unique HTML comment marker representing the phase:\n\n`\u003C!-- engagement-record-{phase}-{ISSUE_ID_OR_PR} -->`\n\nWhere `{phase}` is one of `experience`, `design`, `plan` (at `schema_version: 2`), `orchestration` (at `schema_version: 3`, introduced in v1.3 \u002F #577), or `review` (at `schema_version: 4`, introduced in #655 S3, PR-keyed); and `{ISSUE_ID_OR_PR}` is the GitHub issue ID (for experience\u002Fdesign\u002Fplan\u002Forchestration phases) or PR number (for the `review` phase).\n\nFor example:\n`\u003C!-- engagement-record-design-575 -->`\n`\u003C!-- engagement-record-orchestration-577 -->`\n`\u003C!-- engagement-record-review-42 -->` (PR #42)\n\n## Canonical Schema\n\nThe canonical YAML schema for the engagement-record payload is structured as follows:\n\n```yaml\nschema_version: 3\nphase: orchestration                  # Valid values: experience | design | plan | orchestration | review\ncapture_session: \"normal-orchestration-v3\"  # Session tracking string\nload_bearing_decisions:\n  - decision_id: conductor-scope-classification  # Unique decision identifier\n    classification: load-bearing      # Valid values: load-bearing | routine\n    audit_rationale: \"Rationale...\"   # Short description of the why\n    engineer_choice: \"Choice...\"      # Technical or design choice made\n    teaching_paragraph_excerpt: \"...\" # Excerpt of the decision's context\n    articulation_text: \"...\"          # Short paragraph detailing the engineer's rationale\n    articulation_status: pending      # Valid values: pending | complete | incomplete\n    recommendation_shift_trigger: engineer-pushback # Optional. Valid values: engineer-pushback | new-evidence | classification-re-audit | classification-re-audit-routine\n```\n\n## Persistence Rule\n\nMultiple engagement-record markers may be written to the same issue for a given `(phase, issue)`. The latest-timestamp marker comment (based on GitHub comment `createdAt` field) wins on resume-read. This inherits the standard `SMC-08` \u002F `SMC-17` latest-comment-wins convention.\n\n**Write path (issue #893)**: `skills\u002Fsession-memory-contract\u002Fscripts\u002Fpersist-marker.ps1` (family `engagement-record`, `post-new`, with a payload-only validator adapter that parses the candidate via `Read-EngagementRecords -InMemoryMarkers` before any network write) is the ONLY documented write path for the `experience`, `design`, `plan`, and `orchestration` phases — never a hand-composed `gh issue comment` call. **What that rule buys is a single audited writer — not protection from `updated_at` advancement.** This family is `post-new`, so the write POSTs a fresh comment and advances nothing; the hazard arrives when a write replaces a **whole body** already carrying someone else's marker. See `skills\u002Fsession-memory-contract\u002Freferences\u002Fhandoff-markers.md` § What the write-path rule buys for the write-shape split. **`review`-phase markers (`engagement-record-review-{PR}`) are a known v1 registry gap and stay hand-authored**: the family registry declares `TargetSurface: issue` uniformly, but the `review` phase is PR-keyed, so a `review`-phase write through the script is refused by the surface preflight until that gap is closed.\n\n## Schema Versioning Policy\n\nTooling that reads engagement records MUST throw an error on encountering an unknown `schema_version` value.\n\n- **Additive-field policy**: Within a schema version, new optional fields may be added by writers. Readers MUST ignore unknown optional fields without throwing errors.\n- **Breaking changes**: Any renamed fields, changed enum sets, or new required fields require incrementing the schema version. Readers built against v1.1 throw on v2 markers, and readers built against v1.2 throw on v3 markers — this is intentional; per #576 D4 and #577 D4 the helper is updated in lockstep and `.claude-plugin\u002Fplugin.json` is bumped to invalidate cached older readers. A backward-compatibility guard ensures that `phase: orchestration` requires `schema_version >= 3` to prevent reading orchestration markers with older schemas. A parallel guard requires `phase: review` to have `schema_version >= 4`: `frame-engagement-record-core.ps1` throws if a `review`-phase marker carries `schema_version \u003C 4`.\n\n## Resume-Read Protocol\n\nDownstream\u002Fupstream agents load engagement records at phase startup by calling the helper:\n`Read-EngagementRecords -IssueNumber {ID} [-Phase experience|design|plan|orchestration|review] [-PullRequestNumber {PR}] [-InMemoryMarkers \u003Cstring[]>] [-AcceptLegacy]`\n\nIf a record is returned for a given `decision_id`, the agent activates the `same-decision-resume` skip rule in `skills\u002Fsolution-authoring\u002FSKILL.md` to reuse the captured decision and suppress re-firing the question. For `review`-phase records, pass `-PullRequestNumber {PR}` instead of (or alongside) `-IssueNumber`. See SMC-23.\n\n## articulation_status Transitions\n\n- `pending`: Written by the authoring agent at phase exit.\n- `complete` \u002F `incomplete`: Written by the CE Gate evaluator after assessing the evidence.\n\n> **Initial state**: writers SHOULD emit `articulation_text: \"\"` (empty string) paired with `articulation_status: pending` at phase exit. This is the documented initial state, not an error. The CE Gate (#578) closes the loop by evaluating and transitioning to `complete` or `incomplete`.\n\n## Markdown Bullet ↔ YAML Key Map\n\nWriters and validation tooling MUST enforce the following mapping when translating between the Markdown mirror's bullet fields and the YAML engagement-record's keys:\n\n| Markdown Bullet Field | YAML Payload Key |\n|---|---|\n| `**Classification**` | `classification` |\n| `**Engineer choice**` | `engineer_choice` |\n| `**Audit rationale**` | `audit_rationale` |\n| `**Decision brief excerpt**` | `teaching_paragraph_excerpt` |\n| `**Articulation text**` | `articulation_text` |\n| `**Articulation status**` | `articulation_status` |\n| `**Recommendation shift trigger**` | `recommendation_shift_trigger` |\n\n## Injection Policy\n\nTo prevent Markdown escaping and parsing errors on user-typed fields, writers MUST adhere to the following injection policy:\n\n- Writers MUST use YAML block-scalar `|-` for all multi-line user-typed fields (`audit_rationale`, `articulation_text`, `engineer_choice`).\n- Literal triple-backtick fence lines within those fields are strictly rejected at write time.\n\n## capture_session Policy\n\nThe `capture_session` is a free-form string field that tracks the execution context. The recommended convention is:\n`{trigger}-{phase}-v{major}[.{minor}]`\nWhere:\n\n- `{trigger}` ∈ `normal` | `manual` | `replay`\n- `{phase}` ∈ `experience` | `design` | `plan` | `orchestration` | `review`\n\nReaders MUST NOT reject malformed values, but validation tooling may emit warnings on non-conforming strings.\n\n## Related\n\n- [session-memory-contract](..\u002Fsession-memory-contract\u002FSKILL.md) (`SMC-20`)\n- [solution-authoring](..\u002Fsolution-authoring\u002FSKILL.md) (`same-decision-resume` skip rule)\n- [frame-engagement-record-core](..\u002F..\u002F.github\u002Fscripts\u002Flib\u002Fframe-engagement-record-core.ps1) (`Read-EngagementRecords` helper)\n- **Note on Slug Disambiguation**: The customer-experience skill's prose use of \"named decisions\" refers to the same concept; engagement-record emission persists load-bearing classifications under the same unique slug namespace.\n\n## Gotchas\n\n| Trigger | Gotcha | Fix |\n|---|---|---|\n| Mismatched schema version | An older reader reads a v2 marker and crashes or drops decisions. | Enforce throwing on unknown schema_version and coordinate version bumps across all tools. |\n| `-AcceptLegacy` cross-issue contamination | Before the MF2 fix, `-AcceptLegacy` bypassed the issue-number check, allowing markers from foreign issues to appear in results. | Never pass foreign-issue markers alongside `-AcceptLegacy`; the issue-number check is now always enforced. |\n| Mixed-state issues | Issues that completed earlier phases pre-#576 and later phases post-#576 carry asymmetric resume coverage. | **`schema_version: 1` markers (from #575) ARE read by the helper** — `Read-EngagementRecords` accepts both v1 and v2 markers at parse time, so mixed-state issues with v1 design-phase markers and v2 plan-phase markers resume correctly across both phases. Only purely-pre-#571 markers (lacking `schema_version` entirely) require `-AcceptLegacy` to read; without that switch, they ARE opaque. |\n| Comment-burst halt-on-failure | If engagement-record emission fails after the completion marker is posted, credit-input would still be posted. | Script-enforced (issue #893): the phase-exit burst is one `persist-marker.ps1 -BurstManifest` invocation whose halt-on-first-failure execution stops before credit-input when engagement-record write fails — the completion marker (already landed) stands, the phase is still marked complete, and the script's `⚠️` warning line must be relayed to the terminal. |\n| Markdown summary policy | Markdown bullet lists are human-readable mirrors of the YAML marker payload. | Automation tools must read only the YAML block; the Markdown bullets are for human review\u002Faudits only. |\n| Articulation_text empty-window UX | `articulation_text: \"\"` renders as a blank bullet block in the human-readable Markdown section, which looks like an authoring error. | The Markdown mirror H3 sub-section in agent bodies appends the comment `\u003C!-- CE Gate articulation pending per #578 -->` to clarify that evaluation has not yet occurred. |\n\n## Frame Ports Filled By This Skill\n\nNone. This skill provides supporting methodology and specifications rather than a direct frame port.\n","# Engagement Record Emission\n\nThis skill defines the `SMC-20` engagement-record marker contract and schema used to preserve agentic engagement state across sessions.\n\n## When to Use\n\nUse this skill when an upstream agent (Experience-Owner, Solution-Designer, or Issue-Planner) exits its respective phase. It provides a structured, durable way to record the decisions, classifications, and engineer choices made during the phase, allowing subsequent sessions or downstream agents to skip repetitive questioning on settled decisions.\n\n**Also a writer (#974): the open-for-work conversation.** `skills\u002Fopen-for-work\u002FSKILL.md` records its checkpoint decisions — the worth-it doors, the affirmation gate, and brief approval on the routine arm — into `\u003C!-- engagement-record-experience-{ID} -->` at `schema_version: 2`, because that conversation *is* the experience-replacement for standalone work (#957 D5). It is not an upstream agent, so it is named here explicitly rather than inferred, on the same footing as the `orchestration`-phase carve-out below. **Cumulative-write obligation**: that conversation reaches three checkpoints at different moments, and this family is `post-new` with latest-createdAt-wins on read — so a later marker carrying only the later decision orphans the earlier one. Write one marker at conversation exit carrying every decision, or make each later write carry the full accumulated set.\n\n## When Not to Use\n\nDo not use this skill for runtime code execution, test writing, or PR creation. It is strictly a cognitive-evidence and state-preservation contract. Do not emit these markers in downstream phases. **Exception (v2 carve-out, #655 S3)**: the `review` phase is explicitly permitted to emit `\u003C!-- engagement-record-review-{PR} -->` as a PR comment after the review judge verdict; this marker is PR-keyed (not issue-keyed) and governed by SMC-23.\n\n## Marker Shape\n\nEngagement records are emitted as GitHub issue comments containing a YAML code block wrapped inside a unique HTML comment marker representing the phase:\n\n`\u003C!-- engagement-record-{phase}-{ISSUE_ID_OR_PR} -->`\n\nWhere `{phase}` is one of `experience`, `design`, `plan` (at `schema_version: 2`), `orchestration` (at `schema_version: 3`, introduced in v1.3 \u002F #577), or `review` (at `schema_version: 4`, introduced in #655 S3, PR-keyed); and `{ISSUE_ID_OR_PR}` is the GitHub issue ID (for experience\u002Fdesign\u002Fplan\u002Forchestration phases) or PR number (for the `review` phase).\n\nFor example:\n`\u003C!-- engagement-record-design-575 -->`\n`\u003C!-- engagement-record-orchestration-577 -->`\n`\u003C!-- engagement-record-review-42 -->` (PR #42)\n\n## Canonical Schema\n\nThe canonical YAML schema for the engagement-record payload is structured as follows:\n\n```yaml\nschema_version: 3\nphase: orchestration                  # Valid values: experience | design | plan | orchestration | review\ncapture_session: \"normal-orchestration-v3\"  # Session tracking string\nload_bearing_decisions:\n  - decision_id: conductor-scope-classification  # Unique decision identifier\n    classification: load-bearing      # Valid values: load-bearing | routine\n    audit_rationale: \"Rationale...\"   # Short description of the why\n    engineer_choice: \"Choice...\"      # Technical or design choice made\n    teaching_paragraph_excerpt: \"...\" # Excerpt of the decision's context\n    articulation_text: \"...\"          # Short paragraph detailing the engineer's rationale\n    articulation_status: pending      # Valid values: pending | complete | incomplete\n    recommendation_shift_trigger: engineer-pushback # Optional. Valid values: engineer-pushback | new-evidence | classification-re-audit | classification-re-audit-routine\n```\n\n## Persistence Rule\n\nMultiple engagement-record markers may be written to the same issue for a given `(phase, issue)`. The latest-timestamp marker comment (based on GitHub comment `createdAt` field) wins on resume-read. This inherits the standard `SMC-08` \u002F `SMC-17` latest-comment-wins convention.\n\n**Write path (issue #893)**: `skills\u002Fsession-memory-contract\u002Fscripts\u002Fpersist-marker.ps1` (family `engagement-record`, `post-new`, with a payload-only validator adapter that parses the candidate via `Read-EngagementRecords -InMemoryMarkers` before any network write) is the ONLY documented write path for the `experience`, `design`, `plan`, and `orchestration` phases — never a hand-composed `gh issue comment` call. **What that rule buys is a single audited writer — not protection from `updated_at` advancement.** This family is `post-new`, so the write POSTs a fresh comment and advances nothing; the hazard arrives when a write replaces a **whole body** already carrying someone else's marker. See `skills\u002Fsession-memory-contract\u002Freferences\u002Fhandoff-markers.md` § What the write-path rule buys for the write-shape split. **`review`-phase markers (`engagement-record-review-{PR}`) are a known v1 registry gap and stay hand-authored**: the family registry declares `TargetSurface: issue` uniformly, but the `review` phase is PR-keyed, so a `review`-phase write through the script is refused by the surface preflight until that gap is closed.\n\n## Schema Versioning Policy\n\nTooling that reads engagement records MUST throw an error on encountering an unknown `schema_version` value.\n\n- **Additive-field policy**: Within a schema version, new optional fields may be added by writers. Readers MUST ignore unknown optional fields without throwing errors.\n- **Breaking changes**: Any renamed fields, changed enum sets, or new required fields require incrementing the schema version. Readers built against v1.1 throw on v2 markers, and readers built against v1.2 throw on v3 markers — this is intentional; per #576 D4 and #577 D4 the helper is updated in lockstep and `.claude-plugin\u002Fplugin.json` is bumped to invalidate cached older readers. A backward-compatibility guard ensures that `phase: orchestration` requires `schema_version >= 3` to prevent reading orchestration markers with older schemas. A parallel guard requires `phase: review` to have `schema_version >= 4`: `frame-engagement-record-core.ps1` throws if a `review`-phase marker carries `schema_version \u003C 4`.\n\n## Resume-Read Protocol\n\nDownstream\u002Fupstream agents load engagement records at phase startup by calling the helper:\n`Read-EngagementRecords -IssueNumber {ID} [-Phase experience|design|plan|orchestration|review] [-PullRequestNumber {PR}] [-InMemoryMarkers \u003Cstring[]>] [-AcceptLegacy]`\n\nIf a record is returned for a given `decision_id`, the agent activates the `same-decision-resume` skip rule in `skills\u002Fsolution-authoring\u002FSKILL.md` to reuse the captured decision and suppress re-firing the question. For `review`-phase records, pass `-PullRequestNumber {PR}` instead of (or alongside) `-IssueNumber`. See SMC-23.\n\n## articulation_status Transitions\n\n- `pending`: Written by the authoring agent at phase exit.\n- `complete` \u002F `incomplete`: Written by the CE Gate evaluator after assessing the evidence.\n\n> **Initial state**: writers SHOULD emit `articulation_text: \"\"` (empty string) paired with `articulation_status: pending` at phase exit. This is the documented initial state, not an error. The CE Gate (#578) closes the loop by evaluating and transitioning to `complete` or `incomplete`.\n\n## Markdown Bullet ↔ YAML Key Map\n\nWriters and validation tooling MUST enforce the following mapping when translating between the Markdown mirror's bullet fields and the YAML engagement-record's keys:\n\n| Markdown Bullet Field | YAML Payload Key |\n|---|---|\n| `**Classification**` | `classification` |\n| `**Engineer choice**` | `engineer_choice` |\n| `**Audit rationale**` | `audit_rationale` |\n| `**Decision brief excerpt**` | `teaching_paragraph_excerpt` |\n| `**Articulation text**` | `articulation_text` |\n| `**Articulation status**` | `articulation_status` |\n| `**Recommendation shift trigger**` | `recommendation_shift_trigger` |\n\n## Injection Policy\n\nTo prevent Markdown escaping and parsing errors on user-typed fields, writers MUST adhere to the following injection policy:\n\n- Writers MUST use YAML block-scalar `|-` for all multi-line user-typed fields (`audit_rationale`, `articulation_text`, `engineer_choice`).\n- Literal triple-backtick fence lines within those fields are strictly rejected at write time.\n\n## capture_session Policy\n\nThe `capture_session` is a free-form string field that tracks the execution context. The recommended convention is:\n`{trigger}-{phase}-v{major}[.{minor}]`\nWhere:\n\n- `{trigger}` ∈ `normal` | `manual` | `replay`\n- `{phase}` ∈ `experience` | `design` | `plan` | `orchestration` | `review`\n\nReaders MUST NOT reject malformed values, but validation tooling may emit warnings on non-conforming strings.\n\n## Related\n\n- [session-memory-contract](..\u002Fsession-memory-contract\u002FSKILL.md) (`SMC-20`)\n- [solution-authoring](..\u002Fsolution-authoring\u002FSKILL.md) (`same-decision-resume` skip rule)\n- [frame-engagement-record-core](..\u002F..\u002F.github\u002Fscripts\u002Flib\u002Fframe-engagement-record-core.ps1) (`Read-EngagementRecords` helper)\n- **Note on Slug Disambiguation**: The customer-experience skill's prose use of \"named decisions\" refers to the same concept; engagement-record emission persists load-bearing classifications under the same unique slug namespace.\n\n## Gotchas\n\n| Trigger | Gotcha | Fix |\n|---|---|---|\n| Mismatched schema version | An older reader reads a v2 marker and crashes or drops decisions. | Enforce throwing on unknown schema_version and coordinate version bumps across all tools. |\n| `-AcceptLegacy` cross-issue contamination | Before the MF2 fix, `-AcceptLegacy` bypassed the issue-number check, allowing markers from foreign issues to appear in results. | Never pass foreign-issue markers alongside `-AcceptLegacy`; the issue-number check is now always enforced. |\n| Mixed-state issues | Issues that completed earlier phases pre-#576 and later phases post-#576 carry asymmetric resume coverage. | **`schema_version: 1` markers (from #575) ARE read by the helper** — `Read-EngagementRecords` accepts both v1 and v2 markers at parse time, so mixed-state issues with v1 design-phase markers and v2 plan-phase markers resume correctly across both phases. Only purely-pre-#571 markers (lacking `schema_version` entirely) require `-AcceptLegacy` to read; without that switch, they ARE opaque. |\n| Comment-burst halt-on-failure | If engagement-record emission fails after the completion marker is posted, credit-input would still be posted. | Script-enforced (issue #893): the phase-exit burst is one `persist-marker.ps1 -BurstManifest` invocation whose halt-on-first-failure execution stops before credit-input when engagement-record write fails — the completion marker (already landed) stands, the phase is still marked complete, and the script's `⚠️` warning line must be relayed to the terminal. |\n| Markdown summary policy | Markdown bullet lists are human-readable mirrors of the YAML marker payload. | Automation tools must read only the YAML block; the Markdown bullets are for human review\u002Faudits only. |\n| Articulation_text empty-window UX | `articulation_text: \"\"` renders as a blank bullet block in the human-readable Markdown section, which looks like an authoring error. | The Markdown mirror H3 sub-section in agent bodies appends the comment `\u003C!-- CE Gate articulation pending per #578 -->` to clarify that evaluation has not yet occurred. |\n\n## Frame Ports Filled By This Skill\n\nNone. This skill provides supporting methodology and specifications rather than a direct frame port.\n",{"name":5,"description":7},"main","d5486aacc32c9ca1b07173cfd530f8fe75c29e58","skills\u002Fengagement-record-emission",0,"https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra","https:\u002F\u002Fgithub.com\u002FGrimblaz","2026-08-16 01:12:46",{"source":8,"skillName":5,"commands":42,"prompt":52,"zipAvailable":53},[43,46,49],{"tool":44,"command":45},"npx","npx skills add Grimblaz\u002Fagent-orchestra --skill engagement-record-emission",{"tool":47,"command":48},"bunx","bunx skills add Grimblaz\u002Fagent-orchestra --skill engagement-record-emission",{"tool":50,"command":51},"pnpm","pnpm dlx skills add Grimblaz\u002Fagent-orchestra --skill engagement-record-emission","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fengagement-record-emission\u002FSKILL.md",true,[55],{"path":56,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fconductor-orchestration-record.md","REFERENCE","参考文档",4735,"https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fengagement-record-emission\u002Freferences\u002Fconductor-orchestration-record.md",[62,94,116,132,159,179],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":76,"publishedAt":29},"01M6C48YMJVFHM7DVGG1WWW5RJ","webapp-testing","Playwright end-to-end testing guidance for web apps, with practical patterns and setup steps. Use when creating or improving browser-based E2E coverage, test stability, and CI execution. DO NOT USE FOR: React component tests (use ui-testing), canvas game interaction (use browser-canvas-testing), TDD workflow (use test-driven-development), debugging failures (use systematic-debugging), or randomized property tests (use property-based-testing).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fwebapp-testing\u002FSKILL.md",[68,73],{"categoryNo":69,"categoryName":70,"taxonomyType":71,"taxonomyTypeDesc":72},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",{"categoryNo":74,"categoryName":75,"taxonomyType":71,"taxonomyTypeDesc":72},"01JBSEEDCAT000000000000009","前端开发",[77,78,79,80,81,82,83,21,84,85,86,87,88,89,90,25,91,92,93,27],"testing","webapp testing","webapp","canvas","tdd","workflow","playwright","web","react","component","debugging","development","browser","ci","e2e","game","coverage",{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":29},"01M6C48TNNP2KCSQXV8YK7SK49","verification-before-completion","Evidence-based verification checklist before marking work complete. Use before PRs, releases, marking tickets done, or any \"I'm finished\" declaration; also when naming a baseline commit for a differential suite claim, when designing a mutation campaign or red-run evidence, when making an absence claim, a universal claim, or a one-time demonstration offered as a standing guarantee, when attributing a measurement or claiming a sweep or verification loop is complete, and when stating what a check or a CI suite will do about a change. DO NOT USE FOR: post-merge cleanup or archival (use post-pr-review) or processing GitHub review comments (use code-review-intake).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fverification-before-completion\u002FSKILL.md",[],[101,102,103,104,105,106,22,107,108,23,109,110,90,111,112,113,114,27,28,115],"verification","verification before completion","review","change","commit","merge","cleanup","run","code-review","intake","loop","verification loop","work","check","campaign",{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":122,"publishedAt":29},"01M6C48QZWNTP7JGKD32QZAKKX","validation-methodology","Reusable validation and review methodology for staged validation, failure triage, and prosecution-depth setup. Use when running validation ladders, triaging failures, or executing adversarial review passes. DO NOT USE FOR: CE Gate orchestration, specialist dispatch ownership, or step execution flow (keep those in Code-Conductor).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fvalidation-methodology\u002FSKILL.md",[],[123,124,125,103,126,127,128,129,23,130,25,27,131],"validation","methodology","validation methodology","triage","orchestration","setup","adversarial review","flow","dispatch",{"skillNo":133,"name":134,"nameZh":6,"description":135,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":136,"categories":137,"keywords":146,"publishedAt":29},"01M6C48MCAHG5Z79BJ369CRYRZ","upstream-onboarding","Shared opening-phase protocol for upstream agents (Experience-Owner, Solution-Designer, Issue-Planner) and Code-Conductor when invoked on an existing GitHub issue. Renders a scaled context brief and runs a standards check on inherited work at each phase boundary. Use when a user-invocable agent receives a request referencing an existing GitHub issue and must load solution-authoring and upstream-onboarding at the opening phase. DO NOT USE FOR: subagent dispatches (which already operate within an assessed session context); post-merge review (use post-pr-review); research or non-tree subagents.","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fupstream-onboarding\u002FSKILL.md",[138,141],{"categoryNo":139,"categoryName":140,"taxonomyType":71,"taxonomyTypeDesc":72},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":142,"categoryName":143,"taxonomyType":144,"taxonomyTypeDesc":145},"01JBSEEDCAT000000000000038","人力资源","PROFESSION","适用职业",[147,148,103,149,150,106,151,22,23,152,153,154,155,156,157,113,114,158,26,27],"onboarding","upstream onboarding","github","research","standards check","protocol","brief","boundary","designer","subagent","planner","user",{"skillNo":160,"name":161,"nameZh":6,"description":162,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":163,"categories":164,"keywords":173,"publishedAt":29},"01M6C48GPVE120TGPMD1CS365H","ui-testing","Resilient React component testing strategies focusing on user behavior. Use when writing or reviewing component-level React tests, fixing flaky component tests, or establishing React testing patterns. DO NOT USE FOR: Playwright E2E tests (use webapp-testing), canvas game interaction (use browser-canvas-testing), TDD workflow (use test-driven-development), debugging failures (use systematic-debugging), UI component design (use frontend-design), or randomized property tests (use property-based-testing).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fui-testing\u002FSKILL.md",[165,166,167,170],{"categoryNo":69,"categoryName":70,"taxonomyType":71,"taxonomyTypeDesc":72},{"categoryNo":74,"categoryName":75,"taxonomyType":71,"taxonomyTypeDesc":72},{"categoryNo":168,"categoryName":169,"taxonomyType":71,"taxonomyTypeDesc":72},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":171,"categoryName":172,"taxonomyType":144,"taxonomyTypeDesc":145},"01JBSEEDCAT000000000000032","设计师",[77,174,175,176,177,20,80,82,83,21,85,86,87,88,89,178,91,92,27,79],"ui","ui testing","frontend","design","testing patterns",{"skillNo":180,"name":181,"nameZh":6,"description":182,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":183,"categories":184,"keywords":187,"publishedAt":29},"01M6C48DRJ9XBDR2FW0W37DY53","ui-iteration","Reusable screenshot-driven UI polish workflow for iterative visual refinement. Use when improving layout, hierarchy, spacing, feedback, or Tailwind-based presentation through repeated review-and-adjust passes. DO NOT USE FOR: functional bug fixing, accessibility-only audits, or major redesign strategy work (use Code-Smith or frontend-design).","https:\u002F\u002Fgithub.com\u002FGrimblaz\u002Fagent-orchestra\u002Fblob\u002Fmain\u002Fskills\u002Fui-iteration\u002FSKILL.md",[185,186],{"categoryNo":74,"categoryName":75,"taxonomyType":71,"taxonomyTypeDesc":72},{"categoryNo":171,"categoryName":172,"taxonomyType":144,"taxonomyTypeDesc":145},[174,188,176,177,103,82,189,23,190,191,192,193,194,113,195,196,27,197],"ui iteration","accessibility","screenshot","Tailwind","polish","ui polish","feedback","redesign","strategy","only"]