3.4K skills · sorted by relevance
Review an AI team's work against its acceptance criteria using collected evidence. Match requirements to task results or artifacts, distinguish verified facts from agent claims, and identify uncertainty or missing work before recommending completion.
Run a local check of test evidence and retain source paths in the result. The workflow stops when required evidence is missing, keeping the findings tied to their sources.
Review papers, drafts, and research ideas with evidence-linked feedback focused on improvement. Audit low-stakes assessment rubrics, trace supporting evidence, and check rater agreement and uncertainty without ranking people or recommending consequential decisions.
Retrieve public evidence and linked publications for a supported genetic variant, or look up gene–disease relationships. Preserve sources, conflicting assertions, and explicit search outcomes so qualified professionals can review the findings.
Use a constructive thinking partner to review assumptions, separate evidence from interpretation, and test alternative explanations. Refine conclusions and wording so claims match the available evidence and support decisions under uncertainty.
Manage multi-step financial research with a clear objective, acceptance criteria, and linked evidence. Track findings across turns, audit whether each criterion is supported, and report unresolved gaps while keeping the workflow limited to research.
Coordinate compliance work across multiple standards and regulations. Use a company profile to identify relevant frameworks, map shared controls, prepare mock audits, and build a unified evidence checklist showing where documents can be reused.
Investigate a topic through multiple sources, including primary evidence and counterarguments. Save research materials and notes to files, compare conflicting findings, and produce a cited report that explains conclusions, uncertainty, and limits of the evidence.
Review how evidence can be altered during authorized security exercises on Windows and Linux. Covers logs, timestamps, memory, disk records, and network traces, alongside the signs defenders use to detect manipulation.
Generate, export, load, and verify forensic evidence from GitHub sources. Use when creating verifiable evidence objects from GitHub API, GH Archive, Wayback Machine, local git repositories, or security vendor reports. Handles evidence storage, querying, and re-verification against original sources.
The English summary is being prepared. Showing the original description.
Analysis of Competing Hypotheses. Operationalizes Richards Heuer's CIA-tradition technique for systematically narrowing among multiple hypotheses against evidence. Builds an explicit hypothesis-vs-evidence matrix and focuses on disconfirmation rather than confirmation — the central insight: hypotheses cannot be proven, only disconfirmed; the surviving hypothesis is the one with the least disconfirming evidence. Spawns parallel hypothesizers (in isolation, across hypothesis-generation angles) and parallel evidence-gatherers (across evidence classes), then synthesizes into a matrix, diagnosticity analysis, sensitivity analysis, and falsification milestones. Produces feedback only — no code, no tickets, no artifacts.
The English summary is being prepared. Showing the original description.
Reviews marketing, packaging, web or sustainability text against the environmental-claim rules the user supplies (house policy, regulator guidance extract, industry code) and returns a DRAFT claims review: every environmental claim quoted verbatim with location and family, the evidence the text or evidence pack offers for it with date and scope, the supplied rule passages that bear on each claim, observations as read, and separate question lists for legal and for marketing. No compliance, greenwashing or substantiation verdict. Use when the user asks to "review these green claims", "check this copy against our environmental claims policy", "what evidence do we have for carbon neutral here", "list the sustainability claims in this brochure" or "prepare questions for legal on this eco campaign". Do not use for mapping general factual claims to evidence, use claims-evidence-map instead. Drafts for human review; never approves, authorises or signs off.
The English summary is being prepared. Showing the original description.
diff-time で「影響範囲を repo 全体で調査した証拠」「失敗系を検証した証拠」「外部依存・レート制限・キャッシュ整合を確認した証拠」の欠落を検出する evidence-sufficiency 観点。defect そのものは既存 skill へ委譲し、本 skill は「そのリスク種別を調査した証拠が差分・PR 本文・テストに残っているか」の meta 評価のみを行う。証拠が同一 diff に同梱（テスト・canary・grep 記録）されていれば充足とみなす
The English summary is being prepared. Showing the original description.
Discover, build, verify, and optionally export (HTML, optional PDF) the evidence manifest of one SDD Composy task in evidence_required, routing it toward review_required. Use when acceptance evidence must be assembled or checked — 'juntar as evidências da task', 'gerar o relatório de evidências', 'verify the evidence manifest'. Do NOT use to run tests (sdd-qa), review code (sdd-review), or inspect the trace (sdd-trace).
The English summary is being prepared. Showing the original description.
Audit and refresh OpenClaw maturity scorecards using feature definitions, recorded scores, and QA evidence. Review quality, workflow completeness, and long-term support status, then validate source data and regenerate public documentation.
Investigate performance bottlenecks in REA analysis workflows and improve caching, queries, or shared processing. Compare completed tasks before and after changes while preserving correct results, examined coverage, and access to supporting evidence.
Compare competitors using evidence-backed scores for positioning, brand expression, services, credibility, and other dimensions. Produce consistent profile cards and a two-axis chart showing which rivals occupy the market position your client wants to claim.
Turn interviews, support tickets, surveys, and product context into a single HTML workspace for decisions. Link findings to evidence, compare improvement options, show uncertainty, and outline experiments that help teams choose what to try next.
Pause, resume, or complete an AI team while preserving its work evidence. Check the current state before changes, identify unfinished tasks that prevent run completion, and offer an evidence bundle before closing the work.
Review database migration files before merging changes. It gathers evidence about what the changes do, rollback limits, database locks, and risks from large data volumes, then reports blockers when evidence needed for a decision is missing.
Diagnose problems in Muse Code itself using the installed executable, settings, crash reports, and session evidence. The skill narrows down the likely failing component, summarizes evidence with sensitive information removed, and recommends a safe next action.
Gather scholarly evidence for a scientific manuscript or research brief. Organize supporting and conflicting findings into references, study comparison tables, and claim-to-source maps, with search records and coverage reports that help identify gaps requiring further review.
Prepare constructive review drafts for scientific manuscripts, protocols, preprints, and proposals. Organize checks of claims, evidence, methods, statistics, reproducibility, ethics, figures, and citations into specific comments that a responsible human reviewer can verify and revise.
Turn one scientific paper into a structured, 16-section reading card. Break down its methods, connect claims to experimental evidence, identify limitations, and develop research hypotheses, with source references and clear labels for anything the supplied material cannot support.