[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7C05MNEJV8P914GVPFX81WD":3,"skill-same:01M7C05MNEJV8P914GVPFX81WD":44},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":29,"install":30,"resources":43},"01M7C05MNEJV8P914GVPFX81WD","fidelity-validation",null,"Cross-validates sanitized output specs against raw source specs to detect lost behavioral detail, dropped constants, missing features, or diluted precision. Run AFTER sanitization and AFTER contamination audit passes.","zak-keown\u002Fmoe","zak-keown","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F320717241?v=4","01M7BZFXME0NW6TQX0H7AZM2KS","01M7BZFXN32VR0JN5AATRD53H4",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe\u002Fblob\u002Fmain\u002Fplugins\u002Fmoe-backstory\u002F.cursor-plugin\u002Fskills\u002Ffidelity-validation\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: fidelity-validation\ndescription: Cross-validates sanitized output specs against raw source specs to detect lost behavioral detail, dropped constants, missing features, or diluted precision. Run AFTER sanitization and AFTER contamination audit passes.\n---\n\n# Fidelity Validation\n\nSanitization removes source code identifiers. But aggressive sanitization can also remove behavioral detail that an implementor needs. This skill detects information loss.\n\n## Why This Exists\n\nThe sanitization pass rewrites raw specs into output specs. Each rewrite risks:\n- **Dropped behaviors**: A feature or edge case described in the raw specs that doesn't appear in the output specs\n- **Lost constants**: A numeric threshold, timeout, limit, or size that was accidentally removed or changed during rewriting\n- **Diluted precision**: A specific behavioral rule replaced with vague language (\"the system handles errors\" instead of \"the system retries with configured backoff — N attempts, base delay, maximum delay, and jitter\")\n- **Missing decision trees**: A conditional behavior with multiple branches that was simplified to just the happy path\n- **Dropped error conditions**: Error handling that was described in the raw specs but omitted from the output specs\n- **Feature gaps**: Entire features or sub-features present in the raw specs but missing from the output domain spec\n\n## When to Run\n\nRun this AFTER:\n1. Sanitization (Layer 5) is complete\n2. Contamination audit (Layer 6) passes — output specs are confirmed free of source identifiers\n3. All remediation\u002Frewriting rounds are done\n\nThis is the final quality gate before handing the output to the implementer.\n\n## The Principle\n\n**Every behavioral claim in the raw specs must have a corresponding claim in the output specs — with equal or greater precision.**\n\nThe output specs may use different words (that's the point of sanitization), but it must convey the same behavior. If the raw spec says \"retry N times with a specific base, cap, and jitter\", the output specs must say the same — not just \"the system retries on failure.\"\n\n## Process\n\n### Phase 1: Build the Source Claim Inventory\n\nFor each raw module spec (`workspace\u002Fraw\u002Fspecs\u002Fmodules\u002F*.md`), extract:\n1. **All numeric constants** — timeouts, limits, sizes, counts, thresholds, intervals, percentages\n2. **All behavioral rules** — \"when X happens, the system does Y\"\n3. **All error conditions** — what errors occur and how they're handled\n4. **All state transitions** — state machines, mode changes, lifecycle events\n5. **All decision trees** — if\u002Felse branches, priority orders, cascades\n6. **All features** — distinct capabilities described\n\nWrite this inventory to `workspace\u002Fraw\u002Faudit\u002Ffidelity-inventory.md`.\n\n### Phase 2: Cross-Validate Against Output\n\nFor each item in the source claim inventory, search the corresponding output domain spec(s) for a matching behavioral claim.\n\n**Match criteria:**\n- The output specs describes the same behavior (possibly with different words)\n- Numeric constants match exactly (no rounding, no approximation)\n- Decision branches are all present (not just the happy path)\n- Error conditions are all present\n- State transitions are complete\n\n**Report format per item:**\n\n```\n### [RAW-SPEC: section-name] Claim: \"description of the behavioral claim\"\n- Constant\u002FRule\u002FError\u002FState\u002FDecision\u002FFeature\n- Raw value: [exact value from raw specs]\n- Output location: [file:section where it appears in the output] or MISSING\n- Output value: [exact value from output specs] or N\u002FA\n- Status: MATCH | WEAKENED | MISSING | CHANGED\n- Notes: [if WEAKENED\u002FMISSING\u002FCHANGED, explain what was lost]\n```\n\nStatus definitions:\n- **MATCH**: Output spec conveys the same behavior with the same precision\n- **WEAKENED**: Output spec mentions the behavior but with less precision (e.g., \"retries on failure\" instead of \"retries with specific backoff params\")\n- **MISSING**: Output spec does not mention this behavior at all\n- **CHANGED**: Output spec describes different behavior for the same scenario\n\n### Phase 3: Also Validate Cross-Cutting Artifacts\n\nRepeat Phase 2 for:\n- Journey specs: raw journeys vs output journeys\n- Contracts: raw contracts vs output contracts\n- Protocol specs: raw protocols vs output protocols\n- Test vectors: raw test vectors vs output test vectors (constants must match)\n- Acceptance criteria: raw ACs vs output ACs (criteria must be present)\n\n### Phase 4: Severity Assessment\n\nCategorize all WEAKENED, MISSING, and CHANGED findings:\n\n| Severity | Criteria | Action |\n|----------|----------|--------|\n| **P0-CRITICAL** | A P0 behavior (critical path) is MISSING or CHANGED | Must fix before implementation |\n| **P0-WEAKENED** | A P0 behavior lost precision (constants, branch coverage) | Must fix before implementation |\n| **P1-MISSING** | A P1 behavior is entirely absent from the output | Should fix |\n| **P1-WEAKENED** | A P1 behavior lost precision | Should fix |\n| **P2-MISSING** | A P2\u002FP3 behavior is absent | Advisory — may fix |\n| **P2-WEAKENED** | A P2\u002FP3 behavior lost precision | Advisory — may fix |\n\n### Phase 5: Write Report\n\nWrite the full report to `workspace\u002Fraw\u002Faudit\u002Ffidelity-report.md`.\n\nThe report SHALL include:\n1. **Summary statistics**: total claims checked, MATCH count, WEAKENED count, MISSING count, CHANGED count\n2. **P0-CRITICAL and P0-WEAKENED findings**: full details with exact raw vs output text\n3. **P1 findings**: full details\n4. **P2 findings**: summary only\n5. **Verdict**: PASS (zero P0 issues) or FAIL (any P0 issue exists)\n\nWrite a summary to `workspace\u002Foutput\u002Faudit\u002Ffidelity-summary.md`:\n- Verdict: PASS or FAIL\n- Total claims validated\n- Match rate percentage\n- Number of issues by severity\n- NO details about what was found or what raw specs say — this is an implementer-facing artifact\n\n## Agent Dispatch Pattern\n\nThe fidelity validation is compute-heavy (reading all raw specs AND all output specs). Dispatch as:\n\n1. **Inventory builders** (parallel, one per 4-5 raw module specs): Extract behavioral claims from raw specs → `workspace\u002Fraw\u002Faudit\u002Ffidelity-inventory-batch-N.md`\n\n2. **Cross-validators** (parallel, one per output domain spec): Read the relevant inventory batches + the output domain spec → produce per-domain findings\n\n3. **Aggregator** (sequential): Merge all findings → severity assessment → final report\n\nEach agent is dispatched as `moe-backstory:analyzer` since they need to read both raw\u002F and output\u002F.\n\n## What This Does NOT Check\n\n- Contamination (that's the Layer 6 audit's job)\n- Structural organization (output specs may organize differently than raw — that's fine)\n- Prose quality (rewording is expected)\n- Provenance citations (these are intentionally stripped)\n\n## Relationship to Other Skills\n\n- **spec-sanitization**: Creates the output specs this skill validates\n- **second-pass-review**: Checks for contamination; this skill checks for information loss\n- **validation-methodology**: Defines the acceptance criteria that both raw and output specs must satisfy\n- **behavioral-spec-writing**: Defines what behavioral claims look like\n\nTogether, the audit (Layer 6) and fidelity validation form a two-sided check:\n- Layer 6 asks: \"Is there anything in output\u002F that shouldn't be?\" (contamination)\n- Fidelity validation asks: \"Is there anything NOT in output\u002F that should be?\" (information loss)\n","# Fidelity Validation\n\nSanitization removes source code identifiers. But aggressive sanitization can also remove behavioral detail that an implementor needs. This skill detects information loss.\n\n## Why This Exists\n\nThe sanitization pass rewrites raw specs into output specs. Each rewrite risks:\n- **Dropped behaviors**: A feature or edge case described in the raw specs that doesn't appear in the output specs\n- **Lost constants**: A numeric threshold, timeout, limit, or size that was accidentally removed or changed during rewriting\n- **Diluted precision**: A specific behavioral rule replaced with vague language (\"the system handles errors\" instead of \"the system retries with configured backoff — N attempts, base delay, maximum delay, and jitter\")\n- **Missing decision trees**: A conditional behavior with multiple branches that was simplified to just the happy path\n- **Dropped error conditions**: Error handling that was described in the raw specs but omitted from the output specs\n- **Feature gaps**: Entire features or sub-features present in the raw specs but missing from the output domain spec\n\n## When to Run\n\nRun this AFTER:\n1. Sanitization (Layer 5) is complete\n2. Contamination audit (Layer 6) passes — output specs are confirmed free of source identifiers\n3. All remediation\u002Frewriting rounds are done\n\nThis is the final quality gate before handing the output to the implementer.\n\n## The Principle\n\n**Every behavioral claim in the raw specs must have a corresponding claim in the output specs — with equal or greater precision.**\n\nThe output specs may use different words (that's the point of sanitization), but it must convey the same behavior. If the raw spec says \"retry N times with a specific base, cap, and jitter\", the output specs must say the same — not just \"the system retries on failure.\"\n\n## Process\n\n### Phase 1: Build the Source Claim Inventory\n\nFor each raw module spec (`workspace\u002Fraw\u002Fspecs\u002Fmodules\u002F*.md`), extract:\n1. **All numeric constants** — timeouts, limits, sizes, counts, thresholds, intervals, percentages\n2. **All behavioral rules** — \"when X happens, the system does Y\"\n3. **All error conditions** — what errors occur and how they're handled\n4. **All state transitions** — state machines, mode changes, lifecycle events\n5. **All decision trees** — if\u002Felse branches, priority orders, cascades\n6. **All features** — distinct capabilities described\n\nWrite this inventory to `workspace\u002Fraw\u002Faudit\u002Ffidelity-inventory.md`.\n\n### Phase 2: Cross-Validate Against Output\n\nFor each item in the source claim inventory, search the corresponding output domain spec(s) for a matching behavioral claim.\n\n**Match criteria:**\n- The output specs describes the same behavior (possibly with different words)\n- Numeric constants match exactly (no rounding, no approximation)\n- Decision branches are all present (not just the happy path)\n- Error conditions are all present\n- State transitions are complete\n\n**Report format per item:**\n\n```\n### [RAW-SPEC: section-name] Claim: \"description of the behavioral claim\"\n- Constant\u002FRule\u002FError\u002FState\u002FDecision\u002FFeature\n- Raw value: [exact value from raw specs]\n- Output location: [file:section where it appears in the output] or MISSING\n- Output value: [exact value from output specs] or N\u002FA\n- Status: MATCH | WEAKENED | MISSING | CHANGED\n- Notes: [if WEAKENED\u002FMISSING\u002FCHANGED, explain what was lost]\n```\n\nStatus definitions:\n- **MATCH**: Output spec conveys the same behavior with the same precision\n- **WEAKENED**: Output spec mentions the behavior but with less precision (e.g., \"retries on failure\" instead of \"retries with specific backoff params\")\n- **MISSING**: Output spec does not mention this behavior at all\n- **CHANGED**: Output spec describes different behavior for the same scenario\n\n### Phase 3: Also Validate Cross-Cutting Artifacts\n\nRepeat Phase 2 for:\n- Journey specs: raw journeys vs output journeys\n- Contracts: raw contracts vs output contracts\n- Protocol specs: raw protocols vs output protocols\n- Test vectors: raw test vectors vs output test vectors (constants must match)\n- Acceptance criteria: raw ACs vs output ACs (criteria must be present)\n\n### Phase 4: Severity Assessment\n\nCategorize all WEAKENED, MISSING, and CHANGED findings:\n\n| Severity | Criteria | Action |\n|----------|----------|--------|\n| **P0-CRITICAL** | A P0 behavior (critical path) is MISSING or CHANGED | Must fix before implementation |\n| **P0-WEAKENED** | A P0 behavior lost precision (constants, branch coverage) | Must fix before implementation |\n| **P1-MISSING** | A P1 behavior is entirely absent from the output | Should fix |\n| **P1-WEAKENED** | A P1 behavior lost precision | Should fix |\n| **P2-MISSING** | A P2\u002FP3 behavior is absent | Advisory — may fix |\n| **P2-WEAKENED** | A P2\u002FP3 behavior lost precision | Advisory — may fix |\n\n### Phase 5: Write Report\n\nWrite the full report to `workspace\u002Fraw\u002Faudit\u002Ffidelity-report.md`.\n\nThe report SHALL include:\n1. **Summary statistics**: total claims checked, MATCH count, WEAKENED count, MISSING count, CHANGED count\n2. **P0-CRITICAL and P0-WEAKENED findings**: full details with exact raw vs output text\n3. **P1 findings**: full details\n4. **P2 findings**: summary only\n5. **Verdict**: PASS (zero P0 issues) or FAIL (any P0 issue exists)\n\nWrite a summary to `workspace\u002Foutput\u002Faudit\u002Ffidelity-summary.md`:\n- Verdict: PASS or FAIL\n- Total claims validated\n- Match rate percentage\n- Number of issues by severity\n- NO details about what was found or what raw specs say — this is an implementer-facing artifact\n\n## Agent Dispatch Pattern\n\nThe fidelity validation is compute-heavy (reading all raw specs AND all output specs). Dispatch as:\n\n1. **Inventory builders** (parallel, one per 4-5 raw module specs): Extract behavioral claims from raw specs → `workspace\u002Fraw\u002Faudit\u002Ffidelity-inventory-batch-N.md`\n\n2. **Cross-validators** (parallel, one per output domain spec): Read the relevant inventory batches + the output domain spec → produce per-domain findings\n\n3. **Aggregator** (sequential): Merge all findings → severity assessment → final report\n\nEach agent is dispatched as `moe-backstory:analyzer` since they need to read both raw\u002F and output\u002F.\n\n## What This Does NOT Check\n\n- Contamination (that's the Layer 6 audit's job)\n- Structural organization (output specs may organize differently than raw — that's fine)\n- Prose quality (rewording is expected)\n- Provenance citations (these are intentionally stripped)\n\n## Relationship to Other Skills\n\n- **spec-sanitization**: Creates the output specs this skill validates\n- **second-pass-review**: Checks for contamination; this skill checks for information loss\n- **validation-methodology**: Defines the acceptance criteria that both raw and output specs must satisfy\n- **behavioral-spec-writing**: Defines what behavioral claims look like\n\nTogether, the audit (Layer 6) and fidelity validation form a two-sided check:\n- Layer 6 asks: \"Is there anything in output\u002F that shouldn't be?\" (contamination)\n- Fidelity validation asks: \"Is there anything NOT in output\u002F that should be?\" (information loss)\n",{"name":5,"description":7},"main","728894f3f85fbd4e788f015ddfd72bb6155d5dd5","plugins\u002Fmoe-backstory\u002F.cursor-plugin\u002Fskills\u002Ffidelity-validation",1,"https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe","https:\u002F\u002Fgithub.com\u002Fzak-keown","2026-09-05 06:29:06",{"source":8,"skillName":5,"commands":31,"prompt":41,"zipAvailable":42},[32,35,38],{"tool":33,"command":34},"npx","npx skills add zak-keown\u002Fmoe --skill fidelity-validation",{"tool":36,"command":37},"bunx","bunx skills add zak-keown\u002Fmoe --skill fidelity-validation",{"tool":39,"command":40},"pnpm","pnpm dlx skills add zak-keown\u002Fmoe --skill fidelity-validation","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe\u002Fblob\u002Fmain\u002Fplugins\u002Fmoe-backstory\u002F.cursor-plugin\u002Fskills\u002Ffidelity-validation\u002FSKILL.md",true,[],[45,52,59,66,73,80],{"skillNo":46,"name":47,"nameZh":6,"description":48,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":49,"categories":50,"keywords":51,"publishedAt":18},"01M7C1Z5362Y8NVRZ69HWP0ZD0","write-skill","Use when creating new skills, editing existing skills, or verifying skills work before deployment","https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe\u002Fblob\u002Fmain\u002Fplugins\u002Fmoe\u002Fskills\u002Fwrite-skill\u002FSKILL.md",[],[],{"skillNo":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":18},"01M7C1YZ9966WK9R6ZRPCGWF3G","write-plan","Use when you have a spec or requirements for a multi-step task, before touching code","https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe\u002Fblob\u002Fmain\u002Fplugins\u002Fmoe\u002Fskills\u002Fwrite-plan\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":18},"01M7C1YMR4SJD7BPBBPXZCVGVH","write-clearly","Apply Strunk's timeless writing rules to ANY prose humans will read—documentation, commit messages, error messages, explanations, reports, or UI text. Makes your writing clearer, stronger, and more professional.","https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe\u002Fblob\u002Fmain\u002Fplugins\u002Fmoe\u002Fskills\u002Fwrite-clearly\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":18},"01M7C1YH9805YSM0N20VQAVSXF","windows-vm","Create, manage, or connect to a headless Windows 11 VM running in Docker with SSH access. Use when the user wants to spin up, stop, restart, or SSH into a Windows VM.","https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe\u002Fblob\u002Fmain\u002Fplugins\u002Fmoe\u002Fskills\u002Fwindows-vm\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":18},"01M7C1YD87SCNEJ2C9B8HFR2NQ","verify-completion","Use when about to claim work is complete, fixed, or passing, before committing or creating MRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always","https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe\u002Fblob\u002Fmain\u002Fplugins\u002Fmoe\u002Fskills\u002Fverify-completion\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":18},"01M7C1YARDEHT0TJ9BRG30ZCF4","using-moe","Use when starting any conversation - establishes how to find and use skills, requiring skill invocation before ANY response including clarifying questions","https:\u002F\u002Fgithub.com\u002Fzak-keown\u002Fmoe\u002Fblob\u002Fmain\u002Fplugins\u002Fmoe\u002Fskills\u002Fusing-moe\u002FSKILL.md",[],[]]