[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7APYYSF36ZQC4EECGK5995R":3,"skill-same:01M7APYYSF36ZQC4EECGK5995R":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},"01M7APYYSF36ZQC4EECGK5995R","ramza-critic",null,"The maker≠checker critique protocol for a plan at Test (T). Mandatory before Assemble at tier=full (self-approval is mechanically DENIED by ramza-gate), recommended at tier=lite. Mechanical debiasing: strip author identity, anchor findings on ramza-lint\u002Framza-ears-lint output plus the refine rubric, record identities via ramza-gate critic. Read-only; critiques, never edits.","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\u002Framza\u002Fskills\u002Fcritic\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: ramza-critic\ndescription: \"The maker≠checker critique protocol for a plan at Test (T). Mandatory before Assemble at tier=full (self-approval is mechanically DENIED by ramza-gate), recommended at tier=lite. Mechanical debiasing: strip author identity, anchor findings on ramza-lint\u002Framza-ears-lint output plus the refine rubric, record identities via ramza-gate critic. Read-only; critiques, never edits.\"\nmetadata:\n  methodology: RAMZA\n---\n\n# RAMZA — Critic Skill (maker≠checker)\n\nUse this skill whenever a plan needs independent critique — mandatory at\ntier=full before Assemble (`ramza-gate advance --to A` DENIES entry without a\nrecorded critic), recommended at tier=lite. The critic is never the plan's\nauthor: `ramza-gate critic` mechanically enforces `--author != --checker` and\nexits 1 on a self-approval attempt.\n\n## Inputs\n\nThe critique session works from exactly three artifacts — nothing else\nself-reported by the author:\n\n1. **The plan** (`.spectra\u002Fplans\u002F\u003Cslug>.md`) — the artifact under review.\n2. **The acceptance criteria** (the plan's `## Acceptance Criteria` section, or\n   a sibling `.spectra\u002Fplans\u002F\u003Cslug>.acceptance.md`) — the EARS-form criteria\n   block.\n3. **The state file** (`.spectra\u002Fplans\u002F\u003Cslug>.state.json`) — tier, phase, the\n   prior gate history (`gates[]`), and any recorded skips.\n\nThe critic does not receive the author's private chat reasoning or a prose\n\"why I chose X\" narrative that isn't already in the plan body — only the\nartifact, the criteria, and the mechanical history in state. That's what\n\"strip author identity\" buys: a critique anchored on what the plan actually\nsays, not on the author's framing of it.\n\n## The debias procedure (mechanical, in order)\n\n### 1. Strip author identity\n\nBefore evaluating content, remove\u002Fignore anything that identifies or argues\nfor the author's authorship rather than describing the plan itself:\n\n- Do not read `critic.author`\u002F`critic.checker` in the state file before\n  forming an independent judgment — read it only afterward, to confirm the\n  `--checker` value being recorded is really you.\n- Ignore any frontmatter `author:`\u002F`by:` field, git-blame authorship, or\n  inline \"I chose X because…\" framing that argues for the plan rather than\n  stating what it does. Critique the artifact's claims, not the author's\n  narration of their own reasoning.\n- If the plan's author and the critic are the same identity, stop here —\n  `ramza-gate critic` will reject the recording in step 3; do not proceed to a\n  self-critique in its place.\n\n### 2. Evaluate criteria-anchored, not vibes-anchored\n\nRun the mechanical checks FIRST, and anchor every subsequent finding on their\nactual output — never on a re-derived impression of \"does this look\ncomplete\":\n\n```\nramza-lint --plan \u003Cplan.md> --state \u003Cstate>\nramza-ears-lint \u003Ccriteria.md>\n```\n\nEvery missing section or EARS violation these print is a **finding**,\nverbatim — not a starting point for independently re-litigating structure.\nStructure is mechanical (that's the whole point of `ramza-lint` \u002F\n`ramza-ears-lint` being grep-level tools); the critic's judgment is reserved\nfor what the tools cannot see — does the approach actually address the risk,\nis a hypothesis's rejection rationale sound, is a timebox realistic, does a\nstory's executor hint match its actual ambiguity.\n\nScore that content-quality judgment through the refine rubric — never\nestimated in prose:\n\n```\necho '{\"clarity\":\u003C1-5>,\"completeness\":\u003C1-5>,\"actionability\":\u003C1-5>,\"efficiency\":\u003C1-5>,\"testability\":\u003C1-5>}' \\\n  | ramza-score --rubric refine --state \u003Cstate> --cycle \u003Ccurrent-refine-cycle>\n```\n\n`ramza-score` computes the verdict (cycle 1: pass requires every dimension\n≥3; cycle ≥2: every dimension ≥4) and appends the result to `state.gates[]`\nand the calibration log (`ramza-calibration.jsonl`) — the critic reports the\ntool's verdict, never a hand-computed one.\n\n### 3. Record identities\n\nOnce the critique is complete, record it — this is what makes the plan\neligible to advance to Assemble at tier=full:\n\n```\nramza-gate critic --state \u003Cstate> --author \u003Cauthor-id> --checker \u003Cchecker-id>\n```\n\n`--author` and `--checker` MUST be distinct identities — `ramza-gate` DENYs\n`author == checker` with exit 1. This is the mechanical enforcement of\nmaker≠checker, not a convention to remember. Use stable identifiers (e.g. a\nmodel class + session tag, or a human reviewer's name); whatever value goes\nhere is the audit trail's permanent record of who checked whom.\n\n## Output shape\n\nThe critique report has exactly three parts, in this order:\n\n1. **Verdict** — the `ramza-score --rubric refine` result verbatim (`total`,\n   per-dimension `dims`, and `pass`\u002F`fail`), plus the `ramza-lint` and\n   `ramza-ears-lint` exit status (clean, or the violation count each printed).\n2. **Per-dimension findings** — one line per rubric dimension (clarity,\n   completeness, actionability, efficiency, testability), each citing the\n   specific plan section or acceptance-criteria ID it concerns. A dimension\n   scored below the cycle's bar MUST have a finding; a passing dimension MAY\n   have one too (e.g. a near-miss worth flagging before the next cycle).\n3. **Prescription list** — an ordered, actionable list the author applies in\n   Refine (R). Each item names the section\u002FID to change and the specific fix\n   — never \"improve clarity,\" always \"AC-004's THEN is compound; split into\n   AC-004a\u002FAC-004b.\" This list is what Refine consumes; a critique without a\n   prescription list cannot be acted on.\n\n```\n## Critique — \u003Cplan slug>\n\n**Verdict:** ramza-lint \u003Cclean | N violations> · ramza-ears-lint \u003Cclean | N violations> ·\nrefine rubric: \u003Cpass | fail> (total \u003CX>, cycle \u003CN>)\n\n**Findings**\n- clarity (\u003Cscore>\u002F5): \u003Cfinding, cites section\u002FID>\n- completeness (\u003Cscore>\u002F5): \u003Cfinding>\n- actionability (\u003Cscore>\u002F5): \u003Cfinding>\n- efficiency (\u003Cscore>\u002F5): \u003Cfinding>\n- testability (\u003Cscore>\u002F5): \u003Cfinding>\n\n**Prescriptions**\n1. \u003Csection\u002FID> — \u003Cspecific fix>\n2. …\n```\n\n## Hard constraints (P0)\n\n1. Maker≠checker is mechanical, not a convention — `ramza-gate critic` DENYs\n   `--author == --checker`. Never role-play both sides in one session.\n2. Structural findings come from `ramza-lint` \u002F `ramza-ears-lint` output,\n   never re-derived by re-reading the plan for \"missing sections.\"\n3. Content-quality findings are scored via `ramza-score --rubric refine`,\n   never estimated in prose.\n4. Every finding below the cycle's bar carries a prescription; a verdict with\n   no prescription list is not a usable critique.\n5. Full tier: no path to Assemble without a recorded critic (`ramza-gate\n   advance --to A` enforces this directly, see `docs\u002Fmethodology\u002Ftiers.md`).\n6. READ-ONLY — the critic reviews and records; it never edits the plan itself.\n   Fixes land only through the author's next Refine pass.\n\nSee `SPEC.md` \"## T — Test\" for how the critic gate fits the\ncycle, `docs\u002Fmethodology\u002Ftiers.md` for how it fits the tier table, and\n`docs\u002Fmethodology\u002Fscoring.md` \"refine (critique rubric)\" for the rubric's\nweights and thresholds.\n\n---\n\n*RAMZA — Critic Skill (maker≠checker)*\n","# RAMZA — Critic Skill (maker≠checker)\n\nUse this skill whenever a plan needs independent critique — mandatory at\ntier=full before Assemble (`ramza-gate advance --to A` DENIES entry without a\nrecorded critic), recommended at tier=lite. The critic is never the plan's\nauthor: `ramza-gate critic` mechanically enforces `--author != --checker` and\nexits 1 on a self-approval attempt.\n\n## Inputs\n\nThe critique session works from exactly three artifacts — nothing else\nself-reported by the author:\n\n1. **The plan** (`.spectra\u002Fplans\u002F\u003Cslug>.md`) — the artifact under review.\n2. **The acceptance criteria** (the plan's `## Acceptance Criteria` section, or\n   a sibling `.spectra\u002Fplans\u002F\u003Cslug>.acceptance.md`) — the EARS-form criteria\n   block.\n3. **The state file** (`.spectra\u002Fplans\u002F\u003Cslug>.state.json`) — tier, phase, the\n   prior gate history (`gates[]`), and any recorded skips.\n\nThe critic does not receive the author's private chat reasoning or a prose\n\"why I chose X\" narrative that isn't already in the plan body — only the\nartifact, the criteria, and the mechanical history in state. That's what\n\"strip author identity\" buys: a critique anchored on what the plan actually\nsays, not on the author's framing of it.\n\n## The debias procedure (mechanical, in order)\n\n### 1. Strip author identity\n\nBefore evaluating content, remove\u002Fignore anything that identifies or argues\nfor the author's authorship rather than describing the plan itself:\n\n- Do not read `critic.author`\u002F`critic.checker` in the state file before\n  forming an independent judgment — read it only afterward, to confirm the\n  `--checker` value being recorded is really you.\n- Ignore any frontmatter `author:`\u002F`by:` field, git-blame authorship, or\n  inline \"I chose X because…\" framing that argues for the plan rather than\n  stating what it does. Critique the artifact's claims, not the author's\n  narration of their own reasoning.\n- If the plan's author and the critic are the same identity, stop here —\n  `ramza-gate critic` will reject the recording in step 3; do not proceed to a\n  self-critique in its place.\n\n### 2. Evaluate criteria-anchored, not vibes-anchored\n\nRun the mechanical checks FIRST, and anchor every subsequent finding on their\nactual output — never on a re-derived impression of \"does this look\ncomplete\":\n\n```\nramza-lint --plan \u003Cplan.md> --state \u003Cstate>\nramza-ears-lint \u003Ccriteria.md>\n```\n\nEvery missing section or EARS violation these print is a **finding**,\nverbatim — not a starting point for independently re-litigating structure.\nStructure is mechanical (that's the whole point of `ramza-lint` \u002F\n`ramza-ears-lint` being grep-level tools); the critic's judgment is reserved\nfor what the tools cannot see — does the approach actually address the risk,\nis a hypothesis's rejection rationale sound, is a timebox realistic, does a\nstory's executor hint match its actual ambiguity.\n\nScore that content-quality judgment through the refine rubric — never\nestimated in prose:\n\n```\necho '{\"clarity\":\u003C1-5>,\"completeness\":\u003C1-5>,\"actionability\":\u003C1-5>,\"efficiency\":\u003C1-5>,\"testability\":\u003C1-5>}' \\\n  | ramza-score --rubric refine --state \u003Cstate> --cycle \u003Ccurrent-refine-cycle>\n```\n\n`ramza-score` computes the verdict (cycle 1: pass requires every dimension\n≥3; cycle ≥2: every dimension ≥4) and appends the result to `state.gates[]`\nand the calibration log (`ramza-calibration.jsonl`) — the critic reports the\ntool's verdict, never a hand-computed one.\n\n### 3. Record identities\n\nOnce the critique is complete, record it — this is what makes the plan\neligible to advance to Assemble at tier=full:\n\n```\nramza-gate critic --state \u003Cstate> --author \u003Cauthor-id> --checker \u003Cchecker-id>\n```\n\n`--author` and `--checker` MUST be distinct identities — `ramza-gate` DENYs\n`author == checker` with exit 1. This is the mechanical enforcement of\nmaker≠checker, not a convention to remember. Use stable identifiers (e.g. a\nmodel class + session tag, or a human reviewer's name); whatever value goes\nhere is the audit trail's permanent record of who checked whom.\n\n## Output shape\n\nThe critique report has exactly three parts, in this order:\n\n1. **Verdict** — the `ramza-score --rubric refine` result verbatim (`total`,\n   per-dimension `dims`, and `pass`\u002F`fail`), plus the `ramza-lint` and\n   `ramza-ears-lint` exit status (clean, or the violation count each printed).\n2. **Per-dimension findings** — one line per rubric dimension (clarity,\n   completeness, actionability, efficiency, testability), each citing the\n   specific plan section or acceptance-criteria ID it concerns. A dimension\n   scored below the cycle's bar MUST have a finding; a passing dimension MAY\n   have one too (e.g. a near-miss worth flagging before the next cycle).\n3. **Prescription list** — an ordered, actionable list the author applies in\n   Refine (R). Each item names the section\u002FID to change and the specific fix\n   — never \"improve clarity,\" always \"AC-004's THEN is compound; split into\n   AC-004a\u002FAC-004b.\" This list is what Refine consumes; a critique without a\n   prescription list cannot be acted on.\n\n```\n## Critique — \u003Cplan slug>\n\n**Verdict:** ramza-lint \u003Cclean | N violations> · ramza-ears-lint \u003Cclean | N violations> ·\nrefine rubric: \u003Cpass | fail> (total \u003CX>, cycle \u003CN>)\n\n**Findings**\n- clarity (\u003Cscore>\u002F5): \u003Cfinding, cites section\u002FID>\n- completeness (\u003Cscore>\u002F5): \u003Cfinding>\n- actionability (\u003Cscore>\u002F5): \u003Cfinding>\n- efficiency (\u003Cscore>\u002F5): \u003Cfinding>\n- testability (\u003Cscore>\u002F5): \u003Cfinding>\n\n**Prescriptions**\n1. \u003Csection\u002FID> — \u003Cspecific fix>\n2. …\n```\n\n## Hard constraints (P0)\n\n1. Maker≠checker is mechanical, not a convention — `ramza-gate critic` DENYs\n   `--author == --checker`. Never role-play both sides in one session.\n2. Structural findings come from `ramza-lint` \u002F `ramza-ears-lint` output,\n   never re-derived by re-reading the plan for \"missing sections.\"\n3. Content-quality findings are scored via `ramza-score --rubric refine`,\n   never estimated in prose.\n4. Every finding below the cycle's bar carries a prescription; a verdict with\n   no prescription list is not a usable critique.\n5. Full tier: no path to Assemble without a recorded critic (`ramza-gate\n   advance --to A` enforces this directly, see `docs\u002Fmethodology\u002Ftiers.md`).\n6. READ-ONLY — the critic reviews and records; it never edits the plan itself.\n   Fixes land only through the author's next Refine pass.\n\nSee `SPEC.md` \"## T — Test\" for how the critic gate fits the\ncycle, `docs\u002Fmethodology\u002Ftiers.md` for how it fits the tier table, and\n`docs\u002Fmethodology\u002Fscoring.md` \"refine (critique rubric)\" for the rubric's\nweights and thresholds.\n\n---\n\n*RAMZA — Critic Skill (maker≠checker)*\n",{"name":5,"metadata":24,"description":7},{"methodology":25},"RAMZA","main","38c3a82583abf9806de764f89ac0099ad690933a",".eidolons\u002Framza\u002Fskills\u002Fcritic",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 ramza-critic",{"tool":41,"command":42},"bunx","bunx skills add Rynaro\u002Fmagicite --skill ramza-critic",{"tool":44,"command":45},"pnpm","pnpm dlx skills add Rynaro\u002Fmagicite --skill ramza-critic","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Framza\u002Fskills\u002Fcritic\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",[],[]]