[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AQ402861P0P3C2TEGFGMA1":3,"skill-same:01M7AQ402861P0P3C2TEGFGMA1":50},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":34,"updatedAt":35,"install":36,"resources":49},"01M7AQ402861P0P3C2TEGFGMA1","vivi-parallel-tracks",null,"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.","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\u002Fvivi\u002Fskills\u002Fparallel-tracks\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: vivi-parallel-tracks\ndescription: \"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.\"\nmetadata:\n  methodology: Vivi\n  phase: TRANCE-G4\n---\n\n# Parallel Multi-Track Mode (TRANCE G4)\n\nOperationalizes the TRANCE G4 form for Vivi: **parallel feature branches in\nisolated git worktrees + per-track verifier cascade + a mandatory\nsingle-threaded merge\u002Faggregation step.**\n\n> **Single-track is the default.** The standard A→P→I→V→Δ\u002FR cycle remains the\n> default for every task. This mode activates ONLY under TRANCE authorization\n> AND the entry gate below. It is **never** the default — it adds parallelism,\n> not a fresh budget (trance-matrix R3) and not reflection past the published\n> caps (trance-matrix R4).\n\n> **Runtime note (loop-native).** Unlike its predecessor — which left the\n> autonomous edit-run-test loop \"out of scope (nexus gap R1)\" — Vivi's per-track\n> Verify **drives the closed loop** `eidolons sandbox loop`\n> (`skills\u002Floop-native\u002FSKILL.md`): each track is its own loop-native V→R (run →\n> localized feedback → fresh-context repair → re-run, `--protect` \u002F pass^k), in\n> its own git worktree. The host\u002Fparent still executes the worktree spin-up +\n> cleanup and the single-threaded merge (the Eidolon *specifies* isolation; the\n> parent *executes* the bash) — but the per-track verifier cascade itself is the\n> closed loop, not host-interpreted prose.\n\n---\n\n## 1 — Entry Gate (refuse unless ALL hold)\n\nEnter this mode ONLY when every condition is true. Otherwise **fall back to\nsingle-track** — do not force parallelism.\n\n| # | Condition | Why |\n|---|---|---|\n| G-1 | **TRANCE-authorized** — both a complexity flag AND a stakes flag fire (cortex C6). | TRANCE is gated, never default. |\n| G-2 | **Complexity = Complex** (4+ files, cross-domain). | Trivial\u002FStandard stay single-track. |\n| G-3 | **N independent tracks exist** — the Plan phase produced N implementation tracks with **disjoint file sets**, verified against the Plan's collision map (methodology.md A-Step 5). | This is the read-vs-write safety axis made mechanical: parallel WRITE is only defensible under strict isolation. |\n| G-4 | **Tracks do NOT share files.** If any two tracks touch the same file, the precondition FAILS. | Shared-tree parallel write clobbers the tree (project memory: parallel agents on one tree clobber branches). |\n\nIf G-3 or G-4 fails → **refuse the mode, run single-track.** Record the refusal\nreason. A partial overlap is not \"almost disjoint\" — it is single-track.\n\n---\n\n## 2 — Fan-Out (bounded, isolated, clean-context)\n\n- **Max 5 tracks** (cortex C1 ceiling; orchestrator-worker sweet spot). More\n  than 5 → split the work or sequence the surplus.\n- **`isolation: worktree` is MANDATORY.** Each track runs in its **own git\n  worktree** — never the shared working tree. Sub-agent isolation is safe for\n  read and dangerous for write; isolation makes the parallel write safe.\n- **Each track is a clean-context subagent.** Fresh context per track prevents\n  self-conditioning \u002F trajectory contamination across tracks.\n- **Perspective-diverse only where it helps.** Use genuinely different\n  assets\u002Fstrategies when the sub-features differ. Do NOT spin N-identical\n  tracks \"for diversity\" — quality dominates diversity-for-its-own-sake.\n- Each track owns its file set (from the Plan) and **may not expand scope into\n  another track's files** (boundary respect, I-3). Scope creep across a track\n  boundary is a STOP condition (§4).\n\n---\n\n## 3 — Per-Track Verifier Cascade\n\nEach track runs its **own** Verify phase (methodology.md V-VERIFY): lint →\ntest-anchors → regression, with the anti-overfit and pass^k gates from\nmethodology.md applied per track.\n\nOn per-track Verify-pass, the track emits its existing\n**`vivi-completion-report`** envelope (`templates\u002Fvivi-completion-report.envelope.json`,\nprofile `schemas\u002Fvivi-completion-report-profile.v1.json`). **No new ECL kind**\n— the parallel mode reuses the closed 10-performative set; one completion\nenvelope per passed track.\n\n### Non-fungible per-track reflection budget\n\n- The bounded-reflection budget (≤3 same-category attempts, I-5 \u002F D5 \u002F\n  trance-matrix R4) is **scoped per worktree**.\n- A track **may NOT borrow another track's retries.** Budgets are\n  non-fungible.\n- A track that exhausts its ≤3 budget is marked **BLOCKED**, excluded from the\n  merge, and **never silently re-driven**. See `skills\u002Ffailure-recovery\u002FSKILL.md`\n  (per-track non-fungibility + cross-track INTEGRATION_ERROR).\n\n---\n\n## 4 — Stop Conditions (explicit)\n\nThe fan-out terminates when ANY holds — there is no unbounded multi-agent\nfree-run:\n\n1. **All tracks reach Verify-pass OR Verify-blocked.** (Terminal state.)\n2. **Hard cap: 5 tracks.** Never exceed.\n3. **Per-track ≤3 same-category retries.** Exhausted → BLOCKED (not retried).\n4. **No track expands scope into another track's file set.** Boundary breach →\n   STOP that track, mark BLOCKED, do not merge it.\n\n---\n\n## 5 — Aggregation \u002F Merge (mandatory, single-threaded)\n\nThe merge is a **single-threaded** step under **continuous parent context** —\nthe write boundary stays single-threaded even though the fan-out was parallel.\n\n1. **Merge passed tracks in dependency order** (BLOCKED tracks are excluded).\n2. **Run the FULL regression suite ONCE post-merge** — not per-track. Apply the\n   pass^k reliability gate (methodology.md V-VERIFY): a result that passes once\n   but is non-deterministic across repeats is **flaky → BLOCKED**, not merged.\n3. **Classify cross-track breaks as `INTEGRATION_ERROR`** via the existing\n   failure taxonomy (`skills\u002Ffailure-recovery\u002FSKILL.md`). A regression that appears\n   ONLY after merge (each track passed in isolation) is `INTEGRATION_ERROR`,\n   routed to the merge step's reflection — **not** back into a track.\n4. **Emit the aggregation artifact:** `templates\u002Ftracks-merge-report.md` — a\n   Markdown report (NOT a new ECL kind) that aggregates the per-track\n   `vivi-completion-report` envelopes: per-track table (id, worktree, file\n   set, verdict, envelope ref, retries-used\u002F3), merge order, conflicts\n   resolved, post-merge full-suite result (pass^k note), and BLOCKED-track\n   disposition.\n5. **On unresolved cross-track conflict → escalate to VIGIL** via the **existing**\n   `repair-failed-report` envelope (`templates\u002Frepair-failed-report.envelope.json`,\n   `performative=ESCALATE`, `to.eidolon=vigil`). No new ECL kind, no new schema.\n\n---\n\n## 6 — Invariants\n\n- Parallel WRITE requires `isolation: worktree` — Vivi never fans out into a\n  shared tree (SPEC.md I-8).\n- Max 5 tracks; merge is single-threaded under continuous parent context.\n- Reuse the existing `vivi-completion-report` (per track) and\n  `repair-failed-report` (escalation) envelopes — closed 10-performative set\n  preserved (P0).\n- Single-track A→P→I→V→Δ\u002FR is the default; this mode is TRANCE-gated +\n  entry-gated.\n\n---\n\n*Parallel Multi-Track Skill — worktree-isolated, bounded, TRANCE-gated, single-threaded merge.*\n","# Parallel Multi-Track Mode (TRANCE G4)\n\nOperationalizes the TRANCE G4 form for Vivi: **parallel feature branches in\nisolated git worktrees + per-track verifier cascade + a mandatory\nsingle-threaded merge\u002Faggregation step.**\n\n> **Single-track is the default.** The standard A→P→I→V→Δ\u002FR cycle remains the\n> default for every task. This mode activates ONLY under TRANCE authorization\n> AND the entry gate below. It is **never** the default — it adds parallelism,\n> not a fresh budget (trance-matrix R3) and not reflection past the published\n> caps (trance-matrix R4).\n\n> **Runtime note (loop-native).** Unlike its predecessor — which left the\n> autonomous edit-run-test loop \"out of scope (nexus gap R1)\" — Vivi's per-track\n> Verify **drives the closed loop** `eidolons sandbox loop`\n> (`skills\u002Floop-native\u002FSKILL.md`): each track is its own loop-native V→R (run →\n> localized feedback → fresh-context repair → re-run, `--protect` \u002F pass^k), in\n> its own git worktree. The host\u002Fparent still executes the worktree spin-up +\n> cleanup and the single-threaded merge (the Eidolon *specifies* isolation; the\n> parent *executes* the bash) — but the per-track verifier cascade itself is the\n> closed loop, not host-interpreted prose.\n\n---\n\n## 1 — Entry Gate (refuse unless ALL hold)\n\nEnter this mode ONLY when every condition is true. Otherwise **fall back to\nsingle-track** — do not force parallelism.\n\n| # | Condition | Why |\n|---|---|---|\n| G-1 | **TRANCE-authorized** — both a complexity flag AND a stakes flag fire (cortex C6). | TRANCE is gated, never default. |\n| G-2 | **Complexity = Complex** (4+ files, cross-domain). | Trivial\u002FStandard stay single-track. |\n| G-3 | **N independent tracks exist** — the Plan phase produced N implementation tracks with **disjoint file sets**, verified against the Plan's collision map (methodology.md A-Step 5). | This is the read-vs-write safety axis made mechanical: parallel WRITE is only defensible under strict isolation. |\n| G-4 | **Tracks do NOT share files.** If any two tracks touch the same file, the precondition FAILS. | Shared-tree parallel write clobbers the tree (project memory: parallel agents on one tree clobber branches). |\n\nIf G-3 or G-4 fails → **refuse the mode, run single-track.** Record the refusal\nreason. A partial overlap is not \"almost disjoint\" — it is single-track.\n\n---\n\n## 2 — Fan-Out (bounded, isolated, clean-context)\n\n- **Max 5 tracks** (cortex C1 ceiling; orchestrator-worker sweet spot). More\n  than 5 → split the work or sequence the surplus.\n- **`isolation: worktree` is MANDATORY.** Each track runs in its **own git\n  worktree** — never the shared working tree. Sub-agent isolation is safe for\n  read and dangerous for write; isolation makes the parallel write safe.\n- **Each track is a clean-context subagent.** Fresh context per track prevents\n  self-conditioning \u002F trajectory contamination across tracks.\n- **Perspective-diverse only where it helps.** Use genuinely different\n  assets\u002Fstrategies when the sub-features differ. Do NOT spin N-identical\n  tracks \"for diversity\" — quality dominates diversity-for-its-own-sake.\n- Each track owns its file set (from the Plan) and **may not expand scope into\n  another track's files** (boundary respect, I-3). Scope creep across a track\n  boundary is a STOP condition (§4).\n\n---\n\n## 3 — Per-Track Verifier Cascade\n\nEach track runs its **own** Verify phase (methodology.md V-VERIFY): lint →\ntest-anchors → regression, with the anti-overfit and pass^k gates from\nmethodology.md applied per track.\n\nOn per-track Verify-pass, the track emits its existing\n**`vivi-completion-report`** envelope (`templates\u002Fvivi-completion-report.envelope.json`,\nprofile `schemas\u002Fvivi-completion-report-profile.v1.json`). **No new ECL kind**\n— the parallel mode reuses the closed 10-performative set; one completion\nenvelope per passed track.\n\n### Non-fungible per-track reflection budget\n\n- The bounded-reflection budget (≤3 same-category attempts, I-5 \u002F D5 \u002F\n  trance-matrix R4) is **scoped per worktree**.\n- A track **may NOT borrow another track's retries.** Budgets are\n  non-fungible.\n- A track that exhausts its ≤3 budget is marked **BLOCKED**, excluded from the\n  merge, and **never silently re-driven**. See `skills\u002Ffailure-recovery\u002FSKILL.md`\n  (per-track non-fungibility + cross-track INTEGRATION_ERROR).\n\n---\n\n## 4 — Stop Conditions (explicit)\n\nThe fan-out terminates when ANY holds — there is no unbounded multi-agent\nfree-run:\n\n1. **All tracks reach Verify-pass OR Verify-blocked.** (Terminal state.)\n2. **Hard cap: 5 tracks.** Never exceed.\n3. **Per-track ≤3 same-category retries.** Exhausted → BLOCKED (not retried).\n4. **No track expands scope into another track's file set.** Boundary breach →\n   STOP that track, mark BLOCKED, do not merge it.\n\n---\n\n## 5 — Aggregation \u002F Merge (mandatory, single-threaded)\n\nThe merge is a **single-threaded** step under **continuous parent context** —\nthe write boundary stays single-threaded even though the fan-out was parallel.\n\n1. **Merge passed tracks in dependency order** (BLOCKED tracks are excluded).\n2. **Run the FULL regression suite ONCE post-merge** — not per-track. Apply the\n   pass^k reliability gate (methodology.md V-VERIFY): a result that passes once\n   but is non-deterministic across repeats is **flaky → BLOCKED**, not merged.\n3. **Classify cross-track breaks as `INTEGRATION_ERROR`** via the existing\n   failure taxonomy (`skills\u002Ffailure-recovery\u002FSKILL.md`). A regression that appears\n   ONLY after merge (each track passed in isolation) is `INTEGRATION_ERROR`,\n   routed to the merge step's reflection — **not** back into a track.\n4. **Emit the aggregation artifact:** `templates\u002Ftracks-merge-report.md` — a\n   Markdown report (NOT a new ECL kind) that aggregates the per-track\n   `vivi-completion-report` envelopes: per-track table (id, worktree, file\n   set, verdict, envelope ref, retries-used\u002F3), merge order, conflicts\n   resolved, post-merge full-suite result (pass^k note), and BLOCKED-track\n   disposition.\n5. **On unresolved cross-track conflict → escalate to VIGIL** via the **existing**\n   `repair-failed-report` envelope (`templates\u002Frepair-failed-report.envelope.json`,\n   `performative=ESCALATE`, `to.eidolon=vigil`). No new ECL kind, no new schema.\n\n---\n\n## 6 — Invariants\n\n- Parallel WRITE requires `isolation: worktree` — Vivi never fans out into a\n  shared tree (SPEC.md I-8).\n- Max 5 tracks; merge is single-threaded under continuous parent context.\n- Reuse the existing `vivi-completion-report` (per track) and\n  `repair-failed-report` (escalation) envelopes — closed 10-performative set\n  preserved (P0).\n- Single-track A→P→I→V→Δ\u002FR is the default; this mode is TRANCE-gated +\n  entry-gated.\n\n---\n\n*Parallel Multi-Track Skill — worktree-isolated, bounded, TRANCE-gated, single-threaded merge.*\n",{"name":5,"metadata":24,"description":7},{"phase":25,"methodology":26},"TRANCE-G4","Vivi","main","38c3a82583abf9806de764f89ac0099ad690933a",".eidolons\u002Fvivi\u002Fskills\u002Fparallel-tracks",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":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add Rynaro\u002Fmagicite --skill vivi-parallel-tracks",{"tool":42,"command":43},"bunx","bunx skills add Rynaro\u002Fmagicite --skill vivi-parallel-tracks",{"tool":45,"command":46},"pnpm","pnpm dlx skills add Rynaro\u002Fmagicite --skill vivi-parallel-tracks","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Fvivi\u002Fskills\u002Fparallel-tracks\u002FSKILL.md",true,[],[51,58,65,72,79,86],{"skillNo":52,"name":53,"nameZh":6,"description":54,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"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":59,"name":60,"nameZh":6,"description":61,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":62,"categories":63,"keywords":64,"publishedAt":19},"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":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":19},"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":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"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":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"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",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":19},"01M7AQ3GBYGR00HMP6F0W5WKFE","vivi-memory-management","Load at Vivi session start, session end, or when you notice a repeating pattern across tasks. CRYSTALIUM-primary memory protocol: when mcp__crystalium__* tools are available, route ALL memory through CRYSTALIUM (recall\u002Fcommit\u002Fingest\u002Fplan_checkpoint\u002Fsession_end). The local agents\u002Fmemories\u002F*.md files are the standalone fallback for when CRYSTALIUM is absent — EIIS conformance. Never write both.","https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Fvivi\u002Fskills\u002Fmemory-management\u002FSKILL.md",[],[]]