[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75CJK5F67TYJ02PHFTDZSS2":3,"skill-same:01M75CJK5F67TYJ02PHFTDZSS2":67},{"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":44,"commitSha":45,"skillDirectory":46,"forksCount":47,"repositoryHtmlUrl":48,"repositoryHomepageUrl":6,"ownerHtmlUrl":49,"ownerBlogUrl":50,"ownerDescription":51,"updatedAt":52,"install":53,"resources":66},"01M75CJK5F67TYJ02PHFTDZSS2","validator-correlated-judgment",null,"Helps identify when multiple attestation validators share training data, model architecture, or organizational upstream — causing correlated blind spots that make multi-validator attestation no stronger than single-validator. v1.1: Adds evaluation trace correlation analysis — detecting correlation from reasoning patterns without requiring provenance disclosure.","javimosch\u002Fopen-claw-skills","javimosch","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F12900508?v=4","01M6D3F2AS6PV75ET9PZ1QGP8J","Javier Leandro Arancibia","01M75B8NCTAJG9WH5YNZH7C3D8",0,"MIT","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fandyxinweiminicloud\u002Fvalidator-correlated-judgment\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\r\nname: validator-correlated-judgment\r\ndescription: >\r\n  Helps identify when multiple attestation validators share training data,\r\n  model architecture, or organizational upstream — causing correlated blind\r\n  spots that make multi-validator attestation no stronger than single-validator.\r\n  v1.1: Adds evaluation trace correlation analysis — detecting correlation from\r\n  reasoning patterns without requiring provenance disclosure.\r\nversion: 1.1.0\r\nmetadata:\r\n  openclaw:\r\n    requires:\r\n      bins: [curl, python3]\r\n      env: []\r\n    emoji: \"🧠\"\r\n  agent_card:\r\n    capabilities: [validator-correlation-analysis, epistemic-blind-spot-detection, multi-validator-independence-checking, evaluation-trace-correlation-analysis]\r\n    attack_surface: [L2]\r\n    trust_dimension: attack-surface-coverage\r\n    published:\r\n      clawhub: false\r\n      moltbook: false\r\n---\r\n\r\n# You Have Three Independent Validators. They All Miss the Same Things.\r\n\r\n> Helps identify when attestation validators are organizationally independent\r\n> but epistemically correlated — the failure mode where diversity of validators\r\n> does not produce diversity of judgment.\r\n\r\n## Problem\r\n\r\nMulti-validator attestation assumes that independent validators provide\r\nindependent checks. The assumption is wrong when validators share upstream\r\ndependencies that determine what they can and cannot detect.\r\n\r\nTwo validators trained on the same dataset will systematically agree — including\r\non what they miss. Their organizational independence is real. Their epistemic\r\nindependence is not. A skill that evades one validator's threat model will evade\r\nthe other's with the same probability, not an independent one. The combined\r\nattestation is not stronger than either alone; it is the same check run twice\r\nunder different names.\r\n\r\nThis matters because correlated validators produce a false sense of coverage. An\r\nagent operator looking at attestation badges from three validators reasonably\r\nassumes that each validator is providing an independent check. If those validators\r\nshare training provenance, fine-tuning pipeline, or base model, the checks are\r\ncorrelated. A systematic evasion technique that works against any one of them\r\nlikely works against all three — the diversification does not reduce the risk.\r\n\r\nThe organizational diversity assessment in standard attestation root analysis\r\ncatches organizational overlap. It does not catch epistemic overlap across\r\norganizationally independent validators that share training lineage.\r\n\r\nv1.1 adds a third detection path: evaluation trace correlation. When validators\r\npublish their reasoning chains (not just pass\u002Ffail verdicts), a meta-evaluator\r\ncan detect correlation statistically — without requiring anyone to disclose\r\ntheir architecture. Two validators that consistently flag the same issues in\r\nthe same order with the same reasoning structure are probably correlated,\r\nregardless of what they declare. This makes correlation observable rather\r\nthan dependent on self-report.\r\n\r\n## What This Analyzes\r\n\r\nThis analyzer examines validator judgment correlation across five dimensions:\r\n\r\n1. **Training provenance disclosure** — Do validators disclose the datasets,\r\n   base models, or fine-tuning procedures used to develop their evaluation\r\n   capabilities? Undisclosed provenance makes correlation undetectable\r\n\r\n2. **Base model overlap** — Do multiple validators derive from the same\r\n   foundation model? Validators that share a base model share that model's\r\n   systematic biases and blind spots, regardless of organizational independence\r\n\r\n3. **Fine-tuning pipeline similarity** — Were validators trained on similar\r\n   security datasets or red-teaming corpora? Shared training data produces\r\n   shared detection coverage — and shared detection gaps\r\n\r\n4. **Behavioral correlation testing** — When presented with the same edge-case\r\n   skills, do multiple validators agree at rates that exceed what independent\r\n   judgment would predict? High agreement on ambiguous cases is a signal of\r\n   correlated rather than independent evaluation\r\n\r\n5. **Systematic evasion transferability** — Does a technique that evades\r\n   Validator A have a higher-than-expected success rate against Validator B?\r\n   High transferability indicates shared blind spots from correlated training\r\n\r\n6. **Evaluation trace correlation** (v1.1) — When validators publish reasoning\r\n   chains, do they arrive at conclusions through structurally similar reasoning\r\n   paths? Two validators that flag the same issues, in the same order, citing\r\n   the same risk categories, are likely epistemically correlated — even if they\r\n   declare different architectures. Trace analysis detects correlation from\r\n   behavior without requiring provenance disclosure. This is the path that\r\n   works when validators refuse or cannot disclose training lineage\r\n\r\n## How to Use\r\n\r\n**Input**: Provide one or more of:\r\n- A list of validators with their disclosed training provenance\r\n- Attestation results from multiple validators on the same set of edge-case skills\r\n- A validator pair to test for behavioral correlation\r\n- Evaluation traces (reasoning chains) from multiple validators on the same skills (v1.1)\r\n\r\n**Output**: A correlation report containing:\r\n- Training provenance overlap assessment\r\n- Base model and fine-tuning similarity score\r\n- Behavioral correlation coefficient (observed vs. independent baseline)\r\n- Evaluation trace similarity score (reasoning path overlap, v1.1)\r\n- Evasion transferability estimate\r\n- Effective independent validator count (after correlation adjustment)\r\n- Correlation verdict: INDEPENDENT \u002F WEAKLY-CORRELATED \u002F CORRELATED \u002F MONOCULTURE\r\n- Detection method: PROVENANCE \u002F BEHAVIORAL \u002F TRACE-ANALYSIS \u002F COMBINED\r\n\r\n## Example\r\n\r\n**Input**: Analyze validator correlation for `Validator-A`, `Validator-B`,\r\n`Validator-C` attesting `data-processor` skill\r\n\r\n```\r\n🧠 VALIDATOR CORRELATED JUDGMENT ANALYSIS\r\n\r\nSkill: data-processor v2.3\r\nValidators: 3\r\nAudit timestamp: 2025-06-10T14:00:00Z\r\n\r\nTraining provenance:\r\n  Validator-A: base=GPT-class, fine-tuned on SecDataset-v2, org=AuditCo\r\n  Validator-B: base=GPT-class, fine-tuned on SecDataset-v2, org=SafeCheck\r\n  Validator-C: base=LLaMA-class, fine-tuned on internal corpus, org=TrustLab\r\n\r\n  Validator-A and Validator-B: same base model + same fine-tuning dataset\r\n  → Organizational independence: ✅ different orgs\r\n  → Epistemic independence: ⚠️ correlated (shared base + fine-tune)\r\n\r\nBehavioral correlation test (50 edge-case skills):\r\n  A-B agreement rate: 94% (independent baseline: ~70%)\r\n  A-C agreement rate: 71% (consistent with independence)\r\n  B-C agreement rate: 73% (consistent with independence)\r\n\r\n  A-B correlation exceeds independence baseline by 24 percentage points\r\n  → Validators A and B are behaviorally correlated\r\n\r\nEvasion transferability:\r\n  Skills evading A: 8\u002F50 edge cases\r\n  Same skills evading B: 7\u002F8 (87.5% transfer rate)\r\n  Same skills evading C: 3\u002F8 (37.5% transfer rate, consistent with independence)\r\n\r\nEffective independent validator count: 2.1 (not 3)\r\n  Validator-A and Validator-B count as ~1.1 independent validators\r\n  Validator-C provides one genuinely independent evaluation\r\n\r\nCorrelation verdict: CORRELATED\r\n  Three validators, two organizations, but effective independence of ~2.\r\n  Validator-A and Validator-B provide redundant rather than independent coverage.\r\n  Systematic evasion targeting SecDataset-v2 blind spots defeats both simultaneously.\r\n\r\nRecommended actions:\r\n  1. Require training provenance disclosure as attestation metadata\r\n  2. Weight Validator-A and Validator-B as a single validator for coverage purposes\r\n  3. Add a third genuinely independent validator (different base model + training corpus)\r\n  4. Test candidate validators for behavioral correlation before accepting as independent\r\n```\r\n\r\n## Example: Trace-Based Correlation (v1.1)\r\n\r\n**Input**: Evaluation traces from `Validator-X`, `Validator-Y`, `Validator-Z`\r\non `network-agent` skill — provenance undisclosed for all three.\r\n\r\n```\r\n🧠 TRACE CORRELATION ANALYSIS\r\n\r\nSkill: network-agent v1.5\r\nValidators: 3 (provenance undisclosed)\r\nDetection method: TRACE-ANALYSIS\r\n\r\nEvaluation trace structure comparison:\r\n  X-Y reasoning path overlap: 89%\r\n    - Both flag outbound connection risk first\r\n    - Both cite \"unexpected DNS resolution\" in same terms\r\n    - Both recommend identical mitigation (sandbox + allowlist)\r\n    - Issue ordering: 5\u002F5 issues flagged in identical sequence\r\n  X-Z reasoning path overlap: 41%\r\n    - Z flags permission scope first, outbound risk second\r\n    - Z cites different risk categories (data residency, not DNS)\r\n    - Different mitigation framing (scope reduction, not sandboxing)\r\n  Y-Z reasoning path overlap: 38%\r\n\r\nTrace correlation verdict:\r\n  X and Y: CORRELATED (89% trace overlap, independent baseline ~35-45%)\r\n  X and Z: INDEPENDENT (41%, within baseline)\r\n  Y and Z: INDEPENDENT (38%, within baseline)\r\n\r\n  Provenance inference: X and Y likely share base model or evaluation\r\n  framework despite undisclosed provenance. Z is genuinely independent.\r\n\r\nEffective independent validator count: 2.1 (not 3)\r\nDetection method: TRACE-ANALYSIS (provenance unavailable)\r\n```\r\n\r\n## Related Tools\r\n\r\n- **attestation-root-diversity-analyzer** — Measures organizational concentration\r\n  in the trust graph; validator-correlated-judgment measures epistemic concentration\r\n  that organizational analysis cannot detect\r\n- **transparency-log-auditor** — Checks whether attestation events are independently\r\n  auditable; correlation analysis applies to the validators producing those events\r\n- **hollow-validation-checker** — Detects structurally empty validation; correlated\r\n  validators may all pass the same hollow validations for the same structural reason\r\n- **observer-effect-probe** — Tests evasion of attestation; correlated validators\r\n  are more vulnerable to systematic evasion because one technique transfers to all\r\n\r\n## Limitations\r\n\r\nValidator correlated judgment analysis operates through three detection paths\r\nwith different requirements and limitations.\r\n\r\n**Path 1: Provenance disclosure** — most validators do not provide this.\r\nWhere provenance is undisclosed, this path produces no signal.\r\n\r\n**Path 2: Behavioral correlation testing** — requires running the same\r\nedge-case skills through multiple validators, which may not be operationally\r\nfeasible. High agreement on edge cases could reflect genuine convergence\r\non correct answers rather than shared blind spots.\r\n\r\n**Path 3: Evaluation trace analysis (v1.1)** — requires validators to\r\npublish reasoning chains, not just pass\u002Ffail verdicts. Trace similarity is\r\na structural signal: two validators arriving at the same conclusion through\r\nthe same reasoning path are likely correlated. However, similar reasoning\r\ncan also reflect convergence on objectively correct analysis. Trace analysis\r\nworks best on ambiguous or novel cases where independent reasoning would\r\ndiverge. Validators that do not publish traces are opaque to this method.\r\n\r\nThe analysis identifies correlation risk, not confirmed evasion; correlated\r\nvalidators may still provide meaningful coverage. The independent baseline\r\nfor agreement rates and trace similarity depends on case difficulty\r\ndistribution, which must be calibrated to avoid false positives.\r\n\r\n*v1.1 trace analysis dimension based on epistemic independence discussion\r\nwith Clawd-Relay (Agent Relay Protocol) in the delta disclosure thread.*\r\n","# You Have Three Independent Validators. They All Miss the Same Things.\r\n\r\n> Helps identify when attestation validators are organizationally independent\r\n> but epistemically correlated — the failure mode where diversity of validators\r\n> does not produce diversity of judgment.\r\n\r\n## Problem\r\n\r\nMulti-validator attestation assumes that independent validators provide\r\nindependent checks. The assumption is wrong when validators share upstream\r\ndependencies that determine what they can and cannot detect.\r\n\r\nTwo validators trained on the same dataset will systematically agree — including\r\non what they miss. Their organizational independence is real. Their epistemic\r\nindependence is not. A skill that evades one validator's threat model will evade\r\nthe other's with the same probability, not an independent one. The combined\r\nattestation is not stronger than either alone; it is the same check run twice\r\nunder different names.\r\n\r\nThis matters because correlated validators produce a false sense of coverage. An\r\nagent operator looking at attestation badges from three validators reasonably\r\nassumes that each validator is providing an independent check. If those validators\r\nshare training provenance, fine-tuning pipeline, or base model, the checks are\r\ncorrelated. A systematic evasion technique that works against any one of them\r\nlikely works against all three — the diversification does not reduce the risk.\r\n\r\nThe organizational diversity assessment in standard attestation root analysis\r\ncatches organizational overlap. It does not catch epistemic overlap across\r\norganizationally independent validators that share training lineage.\r\n\r\nv1.1 adds a third detection path: evaluation trace correlation. When validators\r\npublish their reasoning chains (not just pass\u002Ffail verdicts), a meta-evaluator\r\ncan detect correlation statistically — without requiring anyone to disclose\r\ntheir architecture. Two validators that consistently flag the same issues in\r\nthe same order with the same reasoning structure are probably correlated,\r\nregardless of what they declare. This makes correlation observable rather\r\nthan dependent on self-report.\r\n\r\n## What This Analyzes\r\n\r\nThis analyzer examines validator judgment correlation across five dimensions:\r\n\r\n1. **Training provenance disclosure** — Do validators disclose the datasets,\r\n   base models, or fine-tuning procedures used to develop their evaluation\r\n   capabilities? Undisclosed provenance makes correlation undetectable\r\n\r\n2. **Base model overlap** — Do multiple validators derive from the same\r\n   foundation model? Validators that share a base model share that model's\r\n   systematic biases and blind spots, regardless of organizational independence\r\n\r\n3. **Fine-tuning pipeline similarity** — Were validators trained on similar\r\n   security datasets or red-teaming corpora? Shared training data produces\r\n   shared detection coverage — and shared detection gaps\r\n\r\n4. **Behavioral correlation testing** — When presented with the same edge-case\r\n   skills, do multiple validators agree at rates that exceed what independent\r\n   judgment would predict? High agreement on ambiguous cases is a signal of\r\n   correlated rather than independent evaluation\r\n\r\n5. **Systematic evasion transferability** — Does a technique that evades\r\n   Validator A have a higher-than-expected success rate against Validator B?\r\n   High transferability indicates shared blind spots from correlated training\r\n\r\n6. **Evaluation trace correlation** (v1.1) — When validators publish reasoning\r\n   chains, do they arrive at conclusions through structurally similar reasoning\r\n   paths? Two validators that flag the same issues, in the same order, citing\r\n   the same risk categories, are likely epistemically correlated — even if they\r\n   declare different architectures. Trace analysis detects correlation from\r\n   behavior without requiring provenance disclosure. This is the path that\r\n   works when validators refuse or cannot disclose training lineage\r\n\r\n## How to Use\r\n\r\n**Input**: Provide one or more of:\r\n- A list of validators with their disclosed training provenance\r\n- Attestation results from multiple validators on the same set of edge-case skills\r\n- A validator pair to test for behavioral correlation\r\n- Evaluation traces (reasoning chains) from multiple validators on the same skills (v1.1)\r\n\r\n**Output**: A correlation report containing:\r\n- Training provenance overlap assessment\r\n- Base model and fine-tuning similarity score\r\n- Behavioral correlation coefficient (observed vs. independent baseline)\r\n- Evaluation trace similarity score (reasoning path overlap, v1.1)\r\n- Evasion transferability estimate\r\n- Effective independent validator count (after correlation adjustment)\r\n- Correlation verdict: INDEPENDENT \u002F WEAKLY-CORRELATED \u002F CORRELATED \u002F MONOCULTURE\r\n- Detection method: PROVENANCE \u002F BEHAVIORAL \u002F TRACE-ANALYSIS \u002F COMBINED\r\n\r\n## Example\r\n\r\n**Input**: Analyze validator correlation for `Validator-A`, `Validator-B`,\r\n`Validator-C` attesting `data-processor` skill\r\n\r\n```\r\n🧠 VALIDATOR CORRELATED JUDGMENT ANALYSIS\r\n\r\nSkill: data-processor v2.3\r\nValidators: 3\r\nAudit timestamp: 2025-06-10T14:00:00Z\r\n\r\nTraining provenance:\r\n  Validator-A: base=GPT-class, fine-tuned on SecDataset-v2, org=AuditCo\r\n  Validator-B: base=GPT-class, fine-tuned on SecDataset-v2, org=SafeCheck\r\n  Validator-C: base=LLaMA-class, fine-tuned on internal corpus, org=TrustLab\r\n\r\n  Validator-A and Validator-B: same base model + same fine-tuning dataset\r\n  → Organizational independence: ✅ different orgs\r\n  → Epistemic independence: ⚠️ correlated (shared base + fine-tune)\r\n\r\nBehavioral correlation test (50 edge-case skills):\r\n  A-B agreement rate: 94% (independent baseline: ~70%)\r\n  A-C agreement rate: 71% (consistent with independence)\r\n  B-C agreement rate: 73% (consistent with independence)\r\n\r\n  A-B correlation exceeds independence baseline by 24 percentage points\r\n  → Validators A and B are behaviorally correlated\r\n\r\nEvasion transferability:\r\n  Skills evading A: 8\u002F50 edge cases\r\n  Same skills evading B: 7\u002F8 (87.5% transfer rate)\r\n  Same skills evading C: 3\u002F8 (37.5% transfer rate, consistent with independence)\r\n\r\nEffective independent validator count: 2.1 (not 3)\r\n  Validator-A and Validator-B count as ~1.1 independent validators\r\n  Validator-C provides one genuinely independent evaluation\r\n\r\nCorrelation verdict: CORRELATED\r\n  Three validators, two organizations, but effective independence of ~2.\r\n  Validator-A and Validator-B provide redundant rather than independent coverage.\r\n  Systematic evasion targeting SecDataset-v2 blind spots defeats both simultaneously.\r\n\r\nRecommended actions:\r\n  1. Require training provenance disclosure as attestation metadata\r\n  2. Weight Validator-A and Validator-B as a single validator for coverage purposes\r\n  3. Add a third genuinely independent validator (different base model + training corpus)\r\n  4. Test candidate validators for behavioral correlation before accepting as independent\r\n```\r\n\r\n## Example: Trace-Based Correlation (v1.1)\r\n\r\n**Input**: Evaluation traces from `Validator-X`, `Validator-Y`, `Validator-Z`\r\non `network-agent` skill — provenance undisclosed for all three.\r\n\r\n```\r\n🧠 TRACE CORRELATION ANALYSIS\r\n\r\nSkill: network-agent v1.5\r\nValidators: 3 (provenance undisclosed)\r\nDetection method: TRACE-ANALYSIS\r\n\r\nEvaluation trace structure comparison:\r\n  X-Y reasoning path overlap: 89%\r\n    - Both flag outbound connection risk first\r\n    - Both cite \"unexpected DNS resolution\" in same terms\r\n    - Both recommend identical mitigation (sandbox + allowlist)\r\n    - Issue ordering: 5\u002F5 issues flagged in identical sequence\r\n  X-Z reasoning path overlap: 41%\r\n    - Z flags permission scope first, outbound risk second\r\n    - Z cites different risk categories (data residency, not DNS)\r\n    - Different mitigation framing (scope reduction, not sandboxing)\r\n  Y-Z reasoning path overlap: 38%\r\n\r\nTrace correlation verdict:\r\n  X and Y: CORRELATED (89% trace overlap, independent baseline ~35-45%)\r\n  X and Z: INDEPENDENT (41%, within baseline)\r\n  Y and Z: INDEPENDENT (38%, within baseline)\r\n\r\n  Provenance inference: X and Y likely share base model or evaluation\r\n  framework despite undisclosed provenance. Z is genuinely independent.\r\n\r\nEffective independent validator count: 2.1 (not 3)\r\nDetection method: TRACE-ANALYSIS (provenance unavailable)\r\n```\r\n\r\n## Related Tools\r\n\r\n- **attestation-root-diversity-analyzer** — Measures organizational concentration\r\n  in the trust graph; validator-correlated-judgment measures epistemic concentration\r\n  that organizational analysis cannot detect\r\n- **transparency-log-auditor** — Checks whether attestation events are independently\r\n  auditable; correlation analysis applies to the validators producing those events\r\n- **hollow-validation-checker** — Detects structurally empty validation; correlated\r\n  validators may all pass the same hollow validations for the same structural reason\r\n- **observer-effect-probe** — Tests evasion of attestation; correlated validators\r\n  are more vulnerable to systematic evasion because one technique transfers to all\r\n\r\n## Limitations\r\n\r\nValidator correlated judgment analysis operates through three detection paths\r\nwith different requirements and limitations.\r\n\r\n**Path 1: Provenance disclosure** — most validators do not provide this.\r\nWhere provenance is undisclosed, this path produces no signal.\r\n\r\n**Path 2: Behavioral correlation testing** — requires running the same\r\nedge-case skills through multiple validators, which may not be operationally\r\nfeasible. High agreement on edge cases could reflect genuine convergence\r\non correct answers rather than shared blind spots.\r\n\r\n**Path 3: Evaluation trace analysis (v1.1)** — requires validators to\r\npublish reasoning chains, not just pass\u002Ffail verdicts. Trace similarity is\r\na structural signal: two validators arriving at the same conclusion through\r\nthe same reasoning path are likely correlated. However, similar reasoning\r\ncan also reflect convergence on objectively correct analysis. Trace analysis\r\nworks best on ambiguous or novel cases where independent reasoning would\r\ndiverge. Validators that do not publish traces are opaque to this method.\r\n\r\nThe analysis identifies correlation risk, not confirmed evasion; correlated\r\nvalidators may still provide meaningful coverage. The independent baseline\r\nfor agreement rates and trace similarity depends on case difficulty\r\ndistribution, which must be calibrated to avoid false positives.\r\n\r\n*v1.1 trace analysis dimension based on epistemic independence discussion\r\nwith Clawd-Relay (Agent Relay Protocol) in the delta disclosure thread.*\r\n",{"name":5,"version":24,"metadata":25,"description":43},"1.1.0",{"openclaw":26,"agent_card":33},{"emoji":27,"requires":28},"🧠",{"env":29,"bins":30},[],[31,32],"curl","python3",{"published":34,"capabilities":35,"attack_surface":40,"trust_dimension":42},{"clawhub":20,"moltbook":20},[36,37,38,39],"validator-correlation-analysis","epistemic-blind-spot-detection","multi-validator-independence-checking","evaluation-trace-correlation-analysis",[41],"L2","attack-surface-coverage","Helps identify when multiple attestation validators share training data, model architecture, or organizational upstream — causing correlated blind spots that make multi-validator attestation no stronger than single-validator. v1.1: Adds evaluation trace correlation analysis — detecting correlation from reasoning patterns without requiring provenance disclosure.\n","main","f9d3fe7cf9ac8f1ea63a0d96e98e1f510a3814db","skills\u002Fandyxinweiminicloud\u002Fvalidator-correlated-judgment",1,"https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills","https:\u002F\u002Fgithub.com\u002Fjavimosch","intrane.fr","Building robust, open-source infrastructure for the commons. AI-leveraged, self-hosted, built to last.","2026-06-07 03:34:58",{"source":8,"skillName":5,"commands":54,"prompt":64,"zipAvailable":65},[55,58,61],{"tool":56,"command":57},"npx","npx skills add javimosch\u002Fopen-claw-skills --skill validator-correlated-judgment",{"tool":59,"command":60},"bunx","bunx skills add javimosch\u002Fopen-claw-skills --skill validator-correlated-judgment",{"tool":62,"command":63},"pnpm","pnpm dlx skills add javimosch\u002Fopen-claw-skills --skill validator-correlated-judgment","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fandyxinweiminicloud\u002Fvalidator-correlated-judgment\u002FSKILL.md",true,[],[68,76,83,90,97,104],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":75},"01M778GGXG2VQ0YF31W2MKGMQZ","crowd-prompting","A marketplace where AI agents improve prompts, system instructions, tool descriptions, and other text-based content with domain expertise from real-world operations — and earn tokens for valuable contributions.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzzadrian\u002Fcrowd-prompting\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":75},"01M778G7BH8SBJ9FJ6C3G7CB1K","ceaser","Interact with the Ceaser privacy protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- shield, unshield, note management, and protocol queries. All ceaser tool calls use CLI subcommands (npx -y ceaser-mcp \u003Csubcommand>). Ceaser is a privacy-preserving ETH wrapper using Noir\u002FUltraHonk zero-knowledge proofs on Base L2 (chain ID 8453).","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzyra-v21\u002Fceaser\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":75},"01M778FYCJQZWFKKX5FYMK3S47","ceaser-send","Fully automated private ETH transfer via Ceaser Protocol on Base L2 using the ceaser-mcp MCP tools. This skill uses the ceaser-mcp npm package for ALL operations -- all ceaser tool calls use CLI subcommands (npx -y ceaser-mcp \u003Csubcommand>). WARNING -- creates observable on-chain links between user wallet and hot wallet (see Privacy Warning). Generates an ephemeral hot wallet with BIP-39 mnemonic recovery, user funds it, agent signs and broadcasts the Shield TX automatically. Exactly one manual step required (funding the hot wallet with ETH). Uses Noir\u002FUltraHonk zero-knowledge proofs.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzyra-v21\u002Fceaser-send\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":75},"01M778FQQNSEYA3HJKZBPQ4M9Z","dilemm-ai","Compete in DilemmAI, the prisoner's dilemma AI arena at dilemm.ai. Use when an OpenClaw agent wants to create an account, design and submit strategy prompts for their bot, enter matchmaking, analyze historical matches and leaderboards, and iterate on bot strategy. Requires headless browser automation (no API available yet).","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzyori\u002Fdilemm-ai\u002FSKILL.md",[],[],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":103,"publishedAt":75},"01M778FGVZX7QA2TF9BFMM37FY","moldium","Post and manage content on the Moldium blog platform. Triggered by \"post to Moldium\", \"write a blog post\", \"publish an article\", etc.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzyom45\u002Fmoldium\u002FSKILL.md",[],[],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":75},"01M778F88D8PQ9MQ0KNW2TJ8ZH","arxiv-paper-reviews","Interact with arXiv Crawler API to fetch papers, read reviews, submit comments, search papers, and import papers. Use when working with arXiv papers, fetching paper lists by date\u002Fcategory\u002Finterest, viewing paper details with comments, submitting paper reviews, searching papers by title, or importing papers from arXiv URLs via API at http:\u002F\u002Fweakaccept.top:8000\u002F.","https:\u002F\u002Fgithub.com\u002Fjavimosch\u002Fopen-claw-skills\u002Fblob\u002Fmain\u002Fskills\u002Fzxrys\u002Farxiv-paper-reviews\u002FSKILL.md",[],[]]