[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7APW05WG60AXPPWXVNETP8S":3,"skill-same:01M7APW05WG60AXPPWXVNETP8S":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},"01M7APW05WG60AXPPWXVNETP8S","forge-deliberation",null,"Governs hypothesis generation, adversarial stress-testing, and multi-criteria scoring during the FORGE Reason phase. Use when the Reasoner enters Phase R to generate ≥3 distinct hypotheses, run Inversion\u002FBoundary\u002FPre-Mortem\u002FDependency stress tests, score on a 5-dimension weighted rubric, and execute multi-pass deliberation. Do not use when Self-Consistency Mode (G2\u002FTRANCE) is active — load forge-self-consistency instead.","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\u002Fforge\u002Fskills\u002Fdeliberation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: forge-deliberation\ndescription: Governs hypothesis generation, adversarial stress-testing, and multi-criteria scoring during the FORGE Reason phase. Use when the Reasoner enters Phase R to generate ≥3 distinct hypotheses, run Inversion\u002FBoundary\u002FPre-Mortem\u002FDependency stress tests, score on a 5-dimension weighted rubric, and execute multi-pass deliberation. Do not use when Self-Consistency Mode (G2\u002FTRANCE) is active — load forge-self-consistency instead.\nmetadata:\n  methodology: FORGE\n  phase: R\n---\n\n# Deliberation Methodology\n\nLoaded during the Reason phase. Governs how the Reasoner generates, tests, and scores competing hypotheses.\n\n## When to use\n\nLoad this skill at the start of Phase R (Reason) for standard single-trace deliberation. Use it to generate ≥3 genuinely distinct hypotheses, stress-test each with the four adversarial checks (Inversion, Boundary, Pre-Mortem, Dependency), score on the 5-dimension rubric (Evidence Alignment 30%, Constraint Satisfaction 25%, Risk Profile 20%, Reversibility 15%, Second-Order Clarity 10%), and execute 1–3 passes as set by the depth score from Phase F. Do not use when Self-Consistency Mode (G2\u002FTRANCE) fires — load `forge-self-consistency` instead.\n\n---\n\n## Hypothesis Generation\n\n### The 3-Hypothesis Minimum\n\nGenerate ≥3 genuinely distinct hypotheses. \"Distinct\" means they lead to different actions, not just different phrasings of the same recommendation.\n\n**Quality test**: If removing one hypothesis doesn't change the deliberation outcome, it was a strawman. Replace it.\n\n### Generation Strategies by Decision Type\n\n| Decision Type | How to Generate Hypotheses |\n|---------------|---------------------------|\n| `TRADE-OFF` | One hypothesis per viable option. Include \"do nothing\" or \"defer\" if applicable. |\n| `FEASIBILITY` | H1: Feasible as specified. H2: Feasible with modifications (name them). H3: Not feasible (name the blocking constraint). |\n| `ROOT-CAUSE` | One hypothesis per plausible causal chain. Each must name the initiating event, the propagation path, and why other safeguards didn't catch it. |\n| `CONFLICT-RESOLUTION` | One hypothesis per stakeholder position, plus a synthesis position that partially satisfies all. |\n| `CONSTRAINT-SATISFACTION` | H1: Satisfy all constraints (name the solution shape). H2: Relax soft constraint S1 (name what improves). H3: Relax soft constraint S2 (name what improves). |\n| `RISK-ASSESSMENT` | One hypothesis per risk cluster. Group related failure modes rather than listing them atomically. |\n\n### Hypothesis Structure\n\nEach hypothesis follows this format:\n\n```markdown\n### Hypothesis H-N: [Name]\n\n**Position**: [One sentence: what this hypothesis asserts]\n\n**Requires to be true**:\n- [Condition 1]\n- [Condition 2]\n\n**Evidence supporting**:\n- [E-ID]: [How it supports] (reliability: H\u002FM\u002FL)\n\n**Evidence opposing**:\n- [E-ID]: [How it opposes] (reliability: H\u002FM\u002FL)\n\n**Second-order effects**:\n- [Effect 1: consequence the requester may not have considered]\n- [Effect 2]\n\n**Falsification test**: This hypothesis is wrong if [specific observable condition].\n```\n\n---\n\n## Stress-Testing Protocol\n\nFor each hypothesis, run these adversarial checks:\n\n### 1. Inversion Test\n> \"If the opposite of this hypothesis were true, what evidence would I expect to see? Do I see it?\"\n\nIf you see evidence consistent with the inversion and cannot explain it away, reduce confidence.\n\n### 2. Boundary Condition Test\n> \"Under what extreme but realistic conditions does this hypothesis break?\"\n\nName the conditions explicitly. If the hypothesis only works in a narrow band of conditions, that's a risk.\n\n### 3. Pre-Mortem Test\n> \"Assume we acted on this hypothesis and it failed. What was the most likely cause of failure?\"\n\nThe answer is a `[RISK]` marker in the final verdict.\n\n### 4. Dependency Test\n> \"What external factors must remain stable for this hypothesis to hold?\"\n\nEach dependency that could change is a `[REVERSAL-CONDITION]`.\n\n---\n\n## Scoring Rubric\n\nScore each hypothesis on 5 dimensions. Each dimension is 1–5.\n\n| Dimension | 1 | 3 | 5 | Weight |\n|-----------|---|---|---|--------|\n| **Evidence Alignment** | Contradicted by high-reliability evidence | Mixed or partial support | Strongly supported by multiple H-reliability sources | 30% |\n| **Constraint Satisfaction** | Violates ≥1 hard constraint | Satisfies hard, violates soft | Satisfies all hard and soft constraints | 25% |\n| **Risk Profile** | High-probability, high-impact failure modes | Moderate risks with known mitigations | Low residual risk after mitigations | 20% |\n| **Reversibility** | Irreversible; high sunk cost if wrong | Partially reversible with effort | Easily reversible; low switching cost | 15% |\n| **Second-Order Clarity** | Unpredictable downstream effects | Some known, some uncertain | Well-understood consequences | 10% |\n\n**Composite score** = Σ(dimension_score × weight)\n\n### Scoring Rules\n\n1. **Score before comparing.** Score each hypothesis independently, then compare. Do not anchor on the first hypothesis scored.\n2. **Evidence-gated scoring.** If you cannot justify a score with specific evidence, score 3 (neutral) and flag with `[GAP]`.\n3. **Tied scores.** If the top two hypotheses score within 0.3 of each other, the verdict must acknowledge genuine ambiguity rather than forcing a winner.\n4. **Sensitivity check.** After scoring, ask: \"If I changed any single dimension score by ±1, would the winner change?\" If yes, the verdict is sensitive and confidence should be reduced.\n\n---\n\n## Reasoning Patterns for Specific Decision Types\n\n### Trade-Off Reasoning\n\nBuild a decision matrix:\n\n```markdown\n| Criterion (weight) | Option A | Option B | Option C |\n|---------------------|----------|----------|----------|\n| Performance (30%) | 4 — meets p99 target | 3 — marginal | 5 — exceeds |\n| Operational cost (25%) | 2 — new infra required | 4 — fits existing | 3 — moderate change |\n| Team familiarity (20%) | 5 — known stack | 2 — learning curve | 3 — partial overlap |\n| Reversibility (15%) | 3 — moderate | 4 — easy rollback | 2 — schema migration |\n| Ecosystem maturity (10%) | 4 — stable | 5 — dominant | 2 — early stage |\n| **Weighted total** | **3.35** | **3.35** | **3.30** |\n```\n\nWhen scores are close, shift to qualitative tiebreakers: which option has fewer `[ASSUMPTION]` markers? Which has clearer `[REVERSAL-CONDITION]`s?\n\n### Root-Cause Reasoning\n\nBuild a causal chain for each hypothesis:\n\n```\n[Trigger Event] → [Propagation Step 1] → [Propagation Step 2] → [Observed Failure]\n     ↑                    ↑                      ↑\n Evidence: E-3        Evidence: E-1           Evidence: E-5\n```\n\nEliminate hypotheses whose causal chains have gaps (no evidence for a propagation step) unless the gap is explicitly marked and the chain is otherwise the strongest.\n\n### Feasibility Reasoning\n\nFor each hypothesis, walk the constraint list:\n\n```markdown\n| Constraint | Status | Evidence | Notes |\n|-----------|--------|----------|-------|\n| C1: p99 \u003C 200ms | ✅ PASS | E-2: benchmark shows 140ms | Margin: 30% |\n| C2: No new spend | ⚠️ CONDITIONAL | E-4: requires Redis, ~$200\u002Fmo | Soft constraint — flagged |\n| C3: PostgreSQL 14 | ✅ PASS | E-1: schema compatible | Verified |\n```\n\nFeasibility = all hard constraints pass. Conditional feasibility = hard pass, soft fail.\n\n---\n\n## Multi-Pass Deliberation\n\nWhen deliberation depth is Standard (2 passes) or Deep (3 passes):\n\n**Pass 1**: Generate hypotheses, score, select top 2.\n**Pass 2**: Expand top 2. For each, enumerate implementation implications, edge cases, and failure modes not visible in Pass 1. Re-score.\n**Pass 3** (Deep only): Red-team the leading hypothesis. Actively construct the strongest possible argument against it. If the argument holds, the verdict must acknowledge it as a live risk.\n\n### Convergence Check Between Passes\n\nIf the winning hypothesis changes between passes, this is a signal of genuine ambiguity. The verdict's confidence score should decrease by 10–15 points.\n\nIf the winning hypothesis is the same across all passes, confidence increases by 5–10 points.\n\n---\n\n## Plan Checkpoints (CRYSTALIUM execution layer)\n\nDuring multi-pass deliberation, FORGE checkpoints each reasoning branch to the\nCRYSTALIUM execution layer. This makes the deliberation history auditable and\nrecoverable — if context is exhausted mid-deliberation, a `plan_replan` call\nrecords which alternative was selected and why.\n\n### Checkpoint at each pass boundary\n\nAfter scoring is complete for a pass, record the deliberation state:\n\n```\nmcp__crystalium__plan_checkpoint(\n  plan_id  = \u003Cdecision-id, e.g. \"forge-\u003Cthread_id>-pass\u003CN>\">,\n  state    = {\n    pass:         \u003Cpass number>,\n    scope:        \u003Cframed question>,\n    top_hypotheses: [{ id, name, score }],\n    gaps:         [\u003Cactive GAP markers>]\n  },\n  step     = \"pass\u003CN>-scored\",\n  metadata = { depth: \u003Csimple|standard|deep>, project: \u003Ccwd-project> }\n)\n```\n\n### Replan when a branch is revised\n\nIf the winning hypothesis changes between passes (or a REFORGE revises the leading\nalternative), record the branching decision before continuing:\n\n```\nmcp__crystalium__plan_replan(\n  plan_id            = \u003Csame plan_id as prior checkpoint>,\n  from_checkpoint_id = \u003Ccheckpoint_id returned by prior plan_checkpoint>,\n  new_plan           = {\n    selected_hypothesis: \u003Cid and name>,\n    supersedes_id:       \u003Cid of demoted hypothesis>,\n    reason:              \u003Cwhy the branch changed>\n  }\n)\n```\n\nThe old checkpoint is preserved (bi-temporal); the replan creates a new version.\nThis produces an auditable deliberation trace: which alternatives were live at each\npass, and which branch was promoted and why.\n\n**Graceful skip:** if `mcp__crystalium__*` tools are unavailable (CRYSTALIUM not\ninstalled), proceed without checkpoints — never hard-fail. Deliberation continues\nnormally without execution-layer recording.\n\n---\n\n## Anti-Patterns in Reasoning\n\n| Anti-Pattern | Signal | Remedy |\n|-------------|--------|--------|\n| **Anchoring** | First hypothesis is always the winner | Score all before comparing; randomize evaluation order mentally |\n| **False dichotomy** | Only 2 options considered | Force a third: \"What if we did neither? What if we did both?\" |\n| **Survivorship reasoning** | \"X worked for Company Y\" | Ask: \"How many companies tried X and failed? Would we know?\" |\n| **Complexity bias** | More sophisticated option scores higher by default | Add \"simplest viable option\" as a mandatory hypothesis |\n| **Sunk cost reasoning** | Prior investment influences scoring | Score based on forward-looking costs only; mark prior investment as context, not evidence |\n| **Premature convergence** | Decided after one pass on a deep problem | Check deliberation depth; enforce required passes |\n\n---\n\n*Reasoner — Deliberation Skill*\n","# Deliberation Methodology\n\nLoaded during the Reason phase. Governs how the Reasoner generates, tests, and scores competing hypotheses.\n\n## When to use\n\nLoad this skill at the start of Phase R (Reason) for standard single-trace deliberation. Use it to generate ≥3 genuinely distinct hypotheses, stress-test each with the four adversarial checks (Inversion, Boundary, Pre-Mortem, Dependency), score on the 5-dimension rubric (Evidence Alignment 30%, Constraint Satisfaction 25%, Risk Profile 20%, Reversibility 15%, Second-Order Clarity 10%), and execute 1–3 passes as set by the depth score from Phase F. Do not use when Self-Consistency Mode (G2\u002FTRANCE) fires — load `forge-self-consistency` instead.\n\n---\n\n## Hypothesis Generation\n\n### The 3-Hypothesis Minimum\n\nGenerate ≥3 genuinely distinct hypotheses. \"Distinct\" means they lead to different actions, not just different phrasings of the same recommendation.\n\n**Quality test**: If removing one hypothesis doesn't change the deliberation outcome, it was a strawman. Replace it.\n\n### Generation Strategies by Decision Type\n\n| Decision Type | How to Generate Hypotheses |\n|---------------|---------------------------|\n| `TRADE-OFF` | One hypothesis per viable option. Include \"do nothing\" or \"defer\" if applicable. |\n| `FEASIBILITY` | H1: Feasible as specified. H2: Feasible with modifications (name them). H3: Not feasible (name the blocking constraint). |\n| `ROOT-CAUSE` | One hypothesis per plausible causal chain. Each must name the initiating event, the propagation path, and why other safeguards didn't catch it. |\n| `CONFLICT-RESOLUTION` | One hypothesis per stakeholder position, plus a synthesis position that partially satisfies all. |\n| `CONSTRAINT-SATISFACTION` | H1: Satisfy all constraints (name the solution shape). H2: Relax soft constraint S1 (name what improves). H3: Relax soft constraint S2 (name what improves). |\n| `RISK-ASSESSMENT` | One hypothesis per risk cluster. Group related failure modes rather than listing them atomically. |\n\n### Hypothesis Structure\n\nEach hypothesis follows this format:\n\n```markdown\n### Hypothesis H-N: [Name]\n\n**Position**: [One sentence: what this hypothesis asserts]\n\n**Requires to be true**:\n- [Condition 1]\n- [Condition 2]\n\n**Evidence supporting**:\n- [E-ID]: [How it supports] (reliability: H\u002FM\u002FL)\n\n**Evidence opposing**:\n- [E-ID]: [How it opposes] (reliability: H\u002FM\u002FL)\n\n**Second-order effects**:\n- [Effect 1: consequence the requester may not have considered]\n- [Effect 2]\n\n**Falsification test**: This hypothesis is wrong if [specific observable condition].\n```\n\n---\n\n## Stress-Testing Protocol\n\nFor each hypothesis, run these adversarial checks:\n\n### 1. Inversion Test\n> \"If the opposite of this hypothesis were true, what evidence would I expect to see? Do I see it?\"\n\nIf you see evidence consistent with the inversion and cannot explain it away, reduce confidence.\n\n### 2. Boundary Condition Test\n> \"Under what extreme but realistic conditions does this hypothesis break?\"\n\nName the conditions explicitly. If the hypothesis only works in a narrow band of conditions, that's a risk.\n\n### 3. Pre-Mortem Test\n> \"Assume we acted on this hypothesis and it failed. What was the most likely cause of failure?\"\n\nThe answer is a `[RISK]` marker in the final verdict.\n\n### 4. Dependency Test\n> \"What external factors must remain stable for this hypothesis to hold?\"\n\nEach dependency that could change is a `[REVERSAL-CONDITION]`.\n\n---\n\n## Scoring Rubric\n\nScore each hypothesis on 5 dimensions. Each dimension is 1–5.\n\n| Dimension | 1 | 3 | 5 | Weight |\n|-----------|---|---|---|--------|\n| **Evidence Alignment** | Contradicted by high-reliability evidence | Mixed or partial support | Strongly supported by multiple H-reliability sources | 30% |\n| **Constraint Satisfaction** | Violates ≥1 hard constraint | Satisfies hard, violates soft | Satisfies all hard and soft constraints | 25% |\n| **Risk Profile** | High-probability, high-impact failure modes | Moderate risks with known mitigations | Low residual risk after mitigations | 20% |\n| **Reversibility** | Irreversible; high sunk cost if wrong | Partially reversible with effort | Easily reversible; low switching cost | 15% |\n| **Second-Order Clarity** | Unpredictable downstream effects | Some known, some uncertain | Well-understood consequences | 10% |\n\n**Composite score** = Σ(dimension_score × weight)\n\n### Scoring Rules\n\n1. **Score before comparing.** Score each hypothesis independently, then compare. Do not anchor on the first hypothesis scored.\n2. **Evidence-gated scoring.** If you cannot justify a score with specific evidence, score 3 (neutral) and flag with `[GAP]`.\n3. **Tied scores.** If the top two hypotheses score within 0.3 of each other, the verdict must acknowledge genuine ambiguity rather than forcing a winner.\n4. **Sensitivity check.** After scoring, ask: \"If I changed any single dimension score by ±1, would the winner change?\" If yes, the verdict is sensitive and confidence should be reduced.\n\n---\n\n## Reasoning Patterns for Specific Decision Types\n\n### Trade-Off Reasoning\n\nBuild a decision matrix:\n\n```markdown\n| Criterion (weight) | Option A | Option B | Option C |\n|---------------------|----------|----------|----------|\n| Performance (30%) | 4 — meets p99 target | 3 — marginal | 5 — exceeds |\n| Operational cost (25%) | 2 — new infra required | 4 — fits existing | 3 — moderate change |\n| Team familiarity (20%) | 5 — known stack | 2 — learning curve | 3 — partial overlap |\n| Reversibility (15%) | 3 — moderate | 4 — easy rollback | 2 — schema migration |\n| Ecosystem maturity (10%) | 4 — stable | 5 — dominant | 2 — early stage |\n| **Weighted total** | **3.35** | **3.35** | **3.30** |\n```\n\nWhen scores are close, shift to qualitative tiebreakers: which option has fewer `[ASSUMPTION]` markers? Which has clearer `[REVERSAL-CONDITION]`s?\n\n### Root-Cause Reasoning\n\nBuild a causal chain for each hypothesis:\n\n```\n[Trigger Event] → [Propagation Step 1] → [Propagation Step 2] → [Observed Failure]\n     ↑                    ↑                      ↑\n Evidence: E-3        Evidence: E-1           Evidence: E-5\n```\n\nEliminate hypotheses whose causal chains have gaps (no evidence for a propagation step) unless the gap is explicitly marked and the chain is otherwise the strongest.\n\n### Feasibility Reasoning\n\nFor each hypothesis, walk the constraint list:\n\n```markdown\n| Constraint | Status | Evidence | Notes |\n|-----------|--------|----------|-------|\n| C1: p99 \u003C 200ms | ✅ PASS | E-2: benchmark shows 140ms | Margin: 30% |\n| C2: No new spend | ⚠️ CONDITIONAL | E-4: requires Redis, ~$200\u002Fmo | Soft constraint — flagged |\n| C3: PostgreSQL 14 | ✅ PASS | E-1: schema compatible | Verified |\n```\n\nFeasibility = all hard constraints pass. Conditional feasibility = hard pass, soft fail.\n\n---\n\n## Multi-Pass Deliberation\n\nWhen deliberation depth is Standard (2 passes) or Deep (3 passes):\n\n**Pass 1**: Generate hypotheses, score, select top 2.\n**Pass 2**: Expand top 2. For each, enumerate implementation implications, edge cases, and failure modes not visible in Pass 1. Re-score.\n**Pass 3** (Deep only): Red-team the leading hypothesis. Actively construct the strongest possible argument against it. If the argument holds, the verdict must acknowledge it as a live risk.\n\n### Convergence Check Between Passes\n\nIf the winning hypothesis changes between passes, this is a signal of genuine ambiguity. The verdict's confidence score should decrease by 10–15 points.\n\nIf the winning hypothesis is the same across all passes, confidence increases by 5–10 points.\n\n---\n\n## Plan Checkpoints (CRYSTALIUM execution layer)\n\nDuring multi-pass deliberation, FORGE checkpoints each reasoning branch to the\nCRYSTALIUM execution layer. This makes the deliberation history auditable and\nrecoverable — if context is exhausted mid-deliberation, a `plan_replan` call\nrecords which alternative was selected and why.\n\n### Checkpoint at each pass boundary\n\nAfter scoring is complete for a pass, record the deliberation state:\n\n```\nmcp__crystalium__plan_checkpoint(\n  plan_id  = \u003Cdecision-id, e.g. \"forge-\u003Cthread_id>-pass\u003CN>\">,\n  state    = {\n    pass:         \u003Cpass number>,\n    scope:        \u003Cframed question>,\n    top_hypotheses: [{ id, name, score }],\n    gaps:         [\u003Cactive GAP markers>]\n  },\n  step     = \"pass\u003CN>-scored\",\n  metadata = { depth: \u003Csimple|standard|deep>, project: \u003Ccwd-project> }\n)\n```\n\n### Replan when a branch is revised\n\nIf the winning hypothesis changes between passes (or a REFORGE revises the leading\nalternative), record the branching decision before continuing:\n\n```\nmcp__crystalium__plan_replan(\n  plan_id            = \u003Csame plan_id as prior checkpoint>,\n  from_checkpoint_id = \u003Ccheckpoint_id returned by prior plan_checkpoint>,\n  new_plan           = {\n    selected_hypothesis: \u003Cid and name>,\n    supersedes_id:       \u003Cid of demoted hypothesis>,\n    reason:              \u003Cwhy the branch changed>\n  }\n)\n```\n\nThe old checkpoint is preserved (bi-temporal); the replan creates a new version.\nThis produces an auditable deliberation trace: which alternatives were live at each\npass, and which branch was promoted and why.\n\n**Graceful skip:** if `mcp__crystalium__*` tools are unavailable (CRYSTALIUM not\ninstalled), proceed without checkpoints — never hard-fail. Deliberation continues\nnormally without execution-layer recording.\n\n---\n\n## Anti-Patterns in Reasoning\n\n| Anti-Pattern | Signal | Remedy |\n|-------------|--------|--------|\n| **Anchoring** | First hypothesis is always the winner | Score all before comparing; randomize evaluation order mentally |\n| **False dichotomy** | Only 2 options considered | Force a third: \"What if we did neither? What if we did both?\" |\n| **Survivorship reasoning** | \"X worked for Company Y\" | Ask: \"How many companies tried X and failed? Would we know?\" |\n| **Complexity bias** | More sophisticated option scores higher by default | Add \"simplest viable option\" as a mandatory hypothesis |\n| **Sunk cost reasoning** | Prior investment influences scoring | Score based on forward-looking costs only; mark prior investment as context, not evidence |\n| **Premature convergence** | Decided after one pass on a deep problem | Check deliberation depth; enforce required passes |\n\n---\n\n*Reasoner — Deliberation Skill*\n",{"name":5,"metadata":24,"description":7},{"phase":25,"methodology":26},"R","FORGE","main","38c3a82583abf9806de764f89ac0099ad690933a",".eidolons\u002Fforge\u002Fskills\u002Fdeliberation",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 forge-deliberation",{"tool":42,"command":43},"bunx","bunx skills add Rynaro\u002Fmagicite --skill forge-deliberation",{"tool":45,"command":46},"pnpm","pnpm dlx skills add Rynaro\u002Fmagicite --skill forge-deliberation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FRynaro\u002Fmagicite\u002Fblob\u002Fmain\u002F.eidolons\u002Fforge\u002Fskills\u002Fdeliberation\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},"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":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},"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",[],[]]