[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7APXY7WJRQ780MX3A0HSVEH":3,"skill-same:01M7APXY7WJRQ780MX3A0HSVEH":49},{"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":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":6,"ownerHtmlUrl":31,"ownerBlogUrl":32,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M7APXY7WJRQ780MX3A0HSVEH","idg-section-parallel",null,"Operationalizes the G5 gated parallel doc-section synthesis mode for IDG. Use when composing a large document (six or more independent sections) routed at the TRANCE tier; fires dependency-layering, bounded per-layer fan-out (up to five subagents), per-section CHT mini-gate, topological-order parent assembly, and provenance merge. Do NOT use for standard-tier or small documents — sequential composition is always the default.","Rynaro\u002Fmagicite","Rynaro","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F5271543?v=4","01M70CFYFMB316W23THVQAQ42M","Henrique A. Lavezzo","01M7APN2KCVWBNY727T57KR2ZJ",2,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Fidg\u002Fskills\u002Fsection-parallel\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: idg-section-parallel\ndescription: Operationalizes the G5 gated parallel doc-section synthesis mode for IDG. Use when composing a large document (six or more independent sections) routed at the TRANCE tier; fires dependency-layering, bounded per-layer fan-out (up to five subagents), per-section CHT mini-gate, topological-order parent assembly, and provenance merge. Do NOT use for standard-tier or small documents — sequential composition is always the default.\nmetadata:\n  methodology: IDG\n---\n\n# Section-Parallel Synthesis (G5)\n\nOperationalizes the G5 form named in the cortex matrix: **gated parallel doc-section\nsynthesis**. This is the runnable mode behind the sequential \"topological section\norder\" rule in `skills\u002Fcomposition\u002FSKILL.md`. It is **TRANCE-gated and never the default** —\nstandard tier always composes sequentially.\n\nLoad this skill only when the gate below fires. For small documents it is a no-op.\n\n---\n\n## When to use (gate)\n\nActivate section-parallel synthesis only when **all** of these hold:\n\n- The document has **six or more sections** that the skeleton marks as **independent**\n  within a topological layer (no section depends on a sibling in the same layer).\n- The composition is **read-only** — IDG never writes source files, so fan-out carries\n  no write-conflict risk and needs no worktree isolation.\n- The caller (cortex\u002Fhost) routed the task at the **TRANCE** tier.\n\nOtherwise compose sequentially per `skills\u002Fcomposition\u002FSKILL.md`. Small ADRs, runbooks, and\nany document with fewer than six independent sections are an explicit **no-op** — do not\nfan out; the coordination overhead is not repaid.\n\n---\n\n## The five steps\n\n### 1. Dependency-layering\n\nBuild the section dependency graph from the skeleton (as in the Draft phase), then\n**topologically layer** it: layer 0 is every section with no unmet dependency; layer 1\nis every section whose dependencies are all in layer 0; and so on. Sections within a\nlayer are mutually independent by construction and may be drafted concurrently.\n\nRecord the layering in working memory so the parent can assemble in topological order\nlater.\n\n### 2. Bounded per-layer fan-out\n\nFor each layer, in layer order, dispatch **at most five** clean-context per-section\nsubagents — one section per subagent. Each subagent receives **only** the source\nmaterial relevant to its section (the per-section context budget of\n`skills\u002Fcomposition\u002FSKILL.md` still applies: ≤ ~2,000 tokens; summarize and cite beyond that).\n\nA subagent runs in a **clean context** — it sees its section scope, its selected source\nslice, and the already-assembled prior layers as read-only reference. It never sees\nsibling drafts in its own layer (that is what prevents trajectory cross-contamination).\n\nIf a layer has more than five independent sections, batch them: drain five, then the\nnext five, preserving the per-layer cap. Fan-out is **read-only**; no subagent writes to\ndisk and no worktree is created.\n\n### 3. Per-section CHT mini-gate (one revision max)\n\nEach subagent runs a **per-section** CHT mini-gate on its own draft before returning it\n(see `skills\u002Fverification\u002FSKILL.md` — two-granularity CHT). Score the section on Completeness,\nHelpfulness, and Truthfulness:\n\n- All three ≥ 4 → return the section.\n- Any 2–3 → **one** targeted revision pass on the failing dimension, then return with a\n  per-section flag note.\n- Any 1 → return the section stub with an explicit `[GAP]` and the missing-context note;\n  do not loop.\n\n**Hard cap: one revision per section.** No subagent re-drafts more than once.\n\n### 4. Parent assembly (topological-order selection, not averaging)\n\nThe parent collects all returned sections and assembles the document **in topological\norder** (layer 0 first, then layer 1, …). Assembly is **selection, not averaging**: the\nparent takes each section as drafted by its subagent — it does not merge or blend two\ncandidate prose variants into a mean. (IDG drafts each section once per branch; there is\nno multi-candidate vote to average. If the host ever supplies redundant drafts of the\nsame section, the parent **selects** the higher per-section CHT score and discards the\nother; it never concatenates or averages prose.)\n\nWhere two sections make claims that conflict, the parent does not silently pick a winner:\nit emits a `[DISPUTED]` marker presenting both, per `skills\u002Fcomposition\u002FSKILL.md`.\n\n### 5. One document-level coherence pass + provenance merge\n\nAfter assembly the parent runs exactly **one** document-level coherence check (the\nsecond granularity of the CHT gate): cross-section terminology consistency, transition\nbridges between layers, no duplicated or contradictory claims, single provenance block.\n\nThe parent then **merges provenance**: it unions every subagent's per-section source\ncitations, ECL envelope outcomes, and `[GAP]`\u002F`[DISPUTED]` flags into the single\ndocument-level provenance block. Per-section CHT scores roll up to a document-level CHT\nline; the coherence pass contributes the document-level Completeness\u002FHelpfulness check.\n\n**Stop condition (D5):** at most five branches per layer, at most one revision per\nsection, exactly one parent coherence pass. No further iteration. Deliver with flags.\n\n---\n\n## What this mode does NOT change\n\n- **No new write surface.** IDG remains read-only; fan-out is concurrent reads +\n  in-context drafting. No worktree, no file writes by subagents.\n- **No averaging of prose.** Assembly is topological selection; conflicts become\n  `[DISPUTED]`, never a blended mean.\n- **No unbounded loops.** The per-section one-revision cap and the single coherence pass\n  preserve IDG's \"one gate, one revision max\" guarantee at both granularities.\n- **No default change.** Sequential composition stays the standard-tier default; this\n  mode is gated behind the six-independent-section + TRANCE threshold.\n\n---\n\n*Scribe — Section-Parallel Skill*\n","# Section-Parallel Synthesis (G5)\n\nOperationalizes the G5 form named in the cortex matrix: **gated parallel doc-section\nsynthesis**. This is the runnable mode behind the sequential \"topological section\norder\" rule in `skills\u002Fcomposition\u002FSKILL.md`. It is **TRANCE-gated and never the default** —\nstandard tier always composes sequentially.\n\nLoad this skill only when the gate below fires. For small documents it is a no-op.\n\n---\n\n## When to use (gate)\n\nActivate section-parallel synthesis only when **all** of these hold:\n\n- The document has **six or more sections** that the skeleton marks as **independent**\n  within a topological layer (no section depends on a sibling in the same layer).\n- The composition is **read-only** — IDG never writes source files, so fan-out carries\n  no write-conflict risk and needs no worktree isolation.\n- The caller (cortex\u002Fhost) routed the task at the **TRANCE** tier.\n\nOtherwise compose sequentially per `skills\u002Fcomposition\u002FSKILL.md`. Small ADRs, runbooks, and\nany document with fewer than six independent sections are an explicit **no-op** — do not\nfan out; the coordination overhead is not repaid.\n\n---\n\n## The five steps\n\n### 1. Dependency-layering\n\nBuild the section dependency graph from the skeleton (as in the Draft phase), then\n**topologically layer** it: layer 0 is every section with no unmet dependency; layer 1\nis every section whose dependencies are all in layer 0; and so on. Sections within a\nlayer are mutually independent by construction and may be drafted concurrently.\n\nRecord the layering in working memory so the parent can assemble in topological order\nlater.\n\n### 2. Bounded per-layer fan-out\n\nFor each layer, in layer order, dispatch **at most five** clean-context per-section\nsubagents — one section per subagent. Each subagent receives **only** the source\nmaterial relevant to its section (the per-section context budget of\n`skills\u002Fcomposition\u002FSKILL.md` still applies: ≤ ~2,000 tokens; summarize and cite beyond that).\n\nA subagent runs in a **clean context** — it sees its section scope, its selected source\nslice, and the already-assembled prior layers as read-only reference. It never sees\nsibling drafts in its own layer (that is what prevents trajectory cross-contamination).\n\nIf a layer has more than five independent sections, batch them: drain five, then the\nnext five, preserving the per-layer cap. Fan-out is **read-only**; no subagent writes to\ndisk and no worktree is created.\n\n### 3. Per-section CHT mini-gate (one revision max)\n\nEach subagent runs a **per-section** CHT mini-gate on its own draft before returning it\n(see `skills\u002Fverification\u002FSKILL.md` — two-granularity CHT). Score the section on Completeness,\nHelpfulness, and Truthfulness:\n\n- All three ≥ 4 → return the section.\n- Any 2–3 → **one** targeted revision pass on the failing dimension, then return with a\n  per-section flag note.\n- Any 1 → return the section stub with an explicit `[GAP]` and the missing-context note;\n  do not loop.\n\n**Hard cap: one revision per section.** No subagent re-drafts more than once.\n\n### 4. Parent assembly (topological-order selection, not averaging)\n\nThe parent collects all returned sections and assembles the document **in topological\norder** (layer 0 first, then layer 1, …). Assembly is **selection, not averaging**: the\nparent takes each section as drafted by its subagent — it does not merge or blend two\ncandidate prose variants into a mean. (IDG drafts each section once per branch; there is\nno multi-candidate vote to average. If the host ever supplies redundant drafts of the\nsame section, the parent **selects** the higher per-section CHT score and discards the\nother; it never concatenates or averages prose.)\n\nWhere two sections make claims that conflict, the parent does not silently pick a winner:\nit emits a `[DISPUTED]` marker presenting both, per `skills\u002Fcomposition\u002FSKILL.md`.\n\n### 5. One document-level coherence pass + provenance merge\n\nAfter assembly the parent runs exactly **one** document-level coherence check (the\nsecond granularity of the CHT gate): cross-section terminology consistency, transition\nbridges between layers, no duplicated or contradictory claims, single provenance block.\n\nThe parent then **merges provenance**: it unions every subagent's per-section source\ncitations, ECL envelope outcomes, and `[GAP]`\u002F`[DISPUTED]` flags into the single\ndocument-level provenance block. Per-section CHT scores roll up to a document-level CHT\nline; the coherence pass contributes the document-level Completeness\u002FHelpfulness check.\n\n**Stop condition (D5):** at most five branches per layer, at most one revision per\nsection, exactly one parent coherence pass. No further iteration. Deliver with flags.\n\n---\n\n## What this mode does NOT change\n\n- **No new write surface.** IDG remains read-only; fan-out is concurrent reads +\n  in-context drafting. No worktree, no file writes by subagents.\n- **No averaging of prose.** Assembly is topological selection; conflicts become\n  `[DISPUTED]`, never a blended mean.\n- **No unbounded loops.** The per-section one-revision cap and the single coherence pass\n  preserve IDG's \"one gate, one revision max\" guarantee at both granularities.\n- **No default change.** Sequential composition stays the standard-tier default; this\n  mode is gated behind the six-independent-section + TRANCE threshold.\n\n---\n\n*Scribe — Section-Parallel Skill*\n",{"name":5,"metadata":24,"description":7},{"methodology":25},"IDG","main","38c3a82583abf9806de764f89ac0099ad690933a",".eidolons\u002Fidg\u002Fskills\u002Fsection-parallel",0,"https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite","https:\u002F\u002Fgithub.com\u002FRynaro","https:\u002F\u002Fhlavezzo.me","A Code Alchemist with a passion for weaving tales and inscribing transmutation circles in pure code.","2026-10-14 02:35:37",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add Rynaro\u002Fmagicite --skill idg-section-parallel",{"tool":41,"command":42},"bunx","bunx skills add Rynaro\u002Fmagicite --skill idg-section-parallel",{"tool":44,"command":45},"pnpm","pnpm dlx skills add Rynaro\u002Fmagicite --skill idg-section-parallel","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Fidg\u002Fskills\u002Fsection-parallel\u002FSKILL.md",true,[],[50,57,64,71,78,85],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":19},"01M7AQ4GHJQVV8TJBJ8T1V7ZQ9","wine-dxvk-cache-clear","Clear a stale DXVK shader cache that is causing graphical corruption in a\nWine or Proton game. Use when: a game shows corrupted textures or shader\nflicker that started after a driver or DXVK version change.","https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Ftoy-registry\u002Fskills\u002Fwine-dxvk-cache-clear\u002FSKILL.md",[],[],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M7AQ4CY1NHSMY1HCZ91387YY","steam-download-region-fix","Fix a Steam client stuck at very low download speed by changing its\ndownload region. Use when: Steam downloads are far below the local\ninternet speed for every game, not just one.","https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Ftoy-registry\u002Fskills\u002Fsteam-download-region-fix\u002FSKILL.md",[],[],{"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},"01M7AQ49P7Y0CDRSYMPVABX80K","proton-battleye-eac-toggle","Toggle BattlEye or Easy Anti-Cheat runtime support for a Steam Proton game.\nUse when: a multiplayer game refuses to launch with an anti-cheat error under Proton.","https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Ftoy-registry\u002Fskills\u002Fproton-battleye-eac-toggle\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},"01M7AQ45326FAP3X4Q86FK1X4N","vivi-verify-incoming","Load when reading any upstream artefact handed off to Vivi that carries a sibling .envelope.json. BLOCKING per ECL §6.2.2: the orchestrator MUST have verified the envelope's SHA-256 (eidolons verify-envelope --block \u002F eidolons run --verify) and recorded a verify_pass before dispatch. If no verify_pass exists for the message_id, or a verify_fail is present, REFUSE to process the payload and hand back to the orchestrator. Symmetric receiver gate — every Eidolon enforces it identically.","https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Fvivi\u002Fskills\u002Fverify-incoming\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},"01M7AQ402861P0P3C2TEGFGMA1","vivi-parallel-tracks","Load ONLY under TRANCE authorization when the Plan phase yields N independent, non-overlapping implementation tracks (disjoint file sets). Operationalizes the TRANCE G4 form — parallel multi-track implementation in mandatory git-worktree isolation, max 5 clean-context tracks, a per-track verifier cascade, a non-fungible per-track ≤3 reflection budget, explicit stop conditions, and a single-threaded merge\u002Faggregation step. Absent TRANCE gating, Vivi runs exactly as today: single-track A→P→I→V→Δ\u002FR.","https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Fvivi\u002Fskills\u002Fparallel-tracks\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":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M7AQ3RH9BZF9ZMTC7VG5P7PB","vivi-methodology","Full Vivi cycle reference — A(nalyze) → P(lan) → I(mplement) → V(erify) → Δ(Delta)\u002FR(eflect). Use when the task is a non-trivial feature implementation in a brownfield codebase and you need the complete methodology reference: complexity routing, evidence-grounded planning rules, test-anchoring requirements, and failure-escalation thresholds.","https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Fvivi\u002Fskills\u002Fmethodology\u002FSKILL.md",[],[]]