2,492 件 ・ 関連度順
概要と使いどころ
Review and adapt MetaX attention backends, MLA, sparse indexers, cache layouts, and their kernel wrappers against a target vLLM revision and the actually installed MetaX component APIs. Verify dispatch, supported configurations, and GPU numerical behavior. Use for attention adaptations outside vllm_metax/patch/; do not apply monkey-patch headers or patch audit requirements.
日本語の概要は準備中です。原文の説明を表示しています。
Review and adapt vllm_metax/registry registrations, quantization configurations, CustomOps and kernel dispatch against a target vLLM revision and installed MetaX APIs. Verify registration, inherited state, weight layouts and actual backend selection. Use for registry compatibility work, not standalone attention algorithms or monkey patches.
日本語の概要は準備中です。原文の説明を表示しています。
Validate that analytics and marketing tracking fire CORRECTLY: GA4/GTM dataLayer events, Meta/TikTok/LinkedIn pixels, and ad-tech tags. Covers building a tracking plan as the contract, intercepting collect-endpoint beacons and dataLayer.push in Playwright, asserting event name + params + values + timing + de-duplication, Consent Mode v2 gating, CI regression gating, and news-media events (article-view, scroll-depth, paywall). Use when: "test analytics tracking," "GA4 event test," "verify the pixel fires," "dataLayer test," "tracking plan," "Meta Pixel dedup," "scroll-depth tracking test," "gate tracking in CI." Not for: whether tracking is ALLOWED to fire under consent law (GDPR/CMP) — that is compliance-testing; this skill checks the data is CORRECT. SEO meta tags / structured data — out of scope. Related: compliance-testing, playwright-automation, api-testing, qa-project-context.
日本語の概要は準備中です。原文の説明を表示しています。
Validate database integrity, test migrations forward and backward, verify schema constraints, manage seed data, detect migration drift, and identify query performance issues. Covers PostgreSQL, MySQL, MongoDB with Prisma, TypeORM, Drizzle, and SQLAlchemy, plus Testcontainers test databases. Use when: "database test," "migration test," "migration rollback," "rollback test," "data integrity," "SQL test," "schema validation," "seed data," "query performance," "Testcontainers." Not for: synthetic data generation/masking at scale — use test-data-management; Docker/IaC test-environment provisioning — use test-environments; SQL injection — use security-testing. Related: test-data-management, test-environments, security-testing, ci-cd-integration.
日本語の概要は準備中です。原文の説明を表示しています。
Review and assess every issue in the TMDb project board's Backlog column — re-verify each against current main with a read-only fan-out, close the mechanically dead, promote the actionable to Ready with priority, size and execution order, and leave the rest in Backlog naming the decision they need. Use to groom the backlog, before planning a release, or on a schedule. Runs headless: it decides for itself within the rules below.
日本語の概要は準備中です。原文の説明を表示しています。
Run all TMDb unit tests (Swift Testing, the TMDb test plan). Use after code changes and before a PR to verify unit tests pass — delegates the run to the tooling-runner agent (Haiku) and returns total/passed/failed counts plus each failure as Suite/test with file:line. For the live-API suite, use /integration-test.
日本語の概要は準備中です。原文の説明を表示しています。
Fix the currently-failing status checks on the current branch's PR in one sweep — route each failing check to the right diagnosis skill (via a Haiku subagent, escalating to Opus on a repeat failure), apply and verify the fix, commit, and push once — then return a summary. Use standalone when CI is red, or as the check-fixing step invoked by /watch-pr. Repo is adamayoung/TMDb.
日本語の概要は準備中です。原文の説明を表示しています。
Review the working-tree changes (vs main) for correctness, concurrency, architecture, testing, and API/doc issues — following .github/CODE_REVIEW.md — and return a severity-graded report. Scales the machinery to the diff size: a single code-reviewer agent for a small change, or a fan-out-and-verify Workflow for a large/multi-unit one. Produces findings; it does not apply fixes (the caller does).
日本語の概要は準備中です。原文の説明を表示しています。
Audit the knowledge/ base and the .claude/ skills, agents and workflows for staleness, self-contradiction, and drift from their own stated rules — four independent adversarial auditors (two lenses over two trees) verify every load-bearing claim against the current tree, cross-examine each other within their tree, and reach a consensus on if/what needs changing. Use periodically, after a run of deliveries that changed build config, target layout or the skills themselves, or whenever you suspect the docs have aged out of true.
日本語の概要は準備中です。原文の説明を表示しています。
Root cause based one-shot bug fix. debugger diagnosis → requirements.md generation → /execute. /bugfix "error description" Full investigation pipeline: debugger + gap-analyzer + standard verify. QA suggestion after successful fix.
日本語の概要は準備中です。原文の説明を表示しています。
"/blueprint", "blueprint", "task graph", "contract derivation", "execution plan", "plan tasks from requirements", "contract-first planning" Turn requirements.md into an executable blueprint (plan.json + contracts.md). Five phases: Contracts → Tasks → Journeys → Verify Plan → Commit. Sits between /specify and /execute. Scope-adaptive (greenfield → bugfix). Uses hoyeon-cli (plan.json only; requirements.md is read as-is via Read tool).
日本語の概要は準備中です。原文の説明を表示しています。
This skill should be used when the user asks to "analyze session", "evaluate skill execution", "check session logs", provides a session ID with a skill path, or wants to verify that a skill executed correctly in a past session. Post-hoc analysis of Claude Code sessions to validate skill/agent/hook behavior against SKILL.md specifications.
日本語の概要は準備中です。原文の説明を表示しています。
Apply after completing a task, before declaring done. Verify against the real artifact (run the feature, read the actual value, inspect the diff), not a proxy, self-report, or 'it compiles.'
日本語の概要は準備中です。原文の説明を表示しています。
Audits and troubleshoots DNS records for domains including A, AAAA, CNAME, MX, TXT, SPF, DKIM, DMARC, CAA, and NS records. Use when someone needs to verify DNS configuration, debug DNS propagation issues, check email authentication records, or audit domain security. Trigger words: DNS records, dig, nslookup, SPF, DKIM, DMARC, MX records, DNS propagation, nameservers, CAA, domain configuration.
日本語の概要は準備中です。原文の説明を表示しています。
Amazon SES (Simple Email Service) sends and receives email through AWS: transactional and marketing mail, templates, bounce and complaint tracking, and inbound mail processing. Use when a user asks to send email with SES, verify a domain or set up DKIM, move an SES account out of the sandbox, create or bulk-send email templates, add an attachment or unsubscribe header, handle bounces and complaints with SNS, manage the suppression list, or store incoming mail in S3 with receipt rules. Covers the SES v2 API (aws sesv2, boto3 sesv2).
日本語の概要は準備中です。原文の説明を表示しています。
Test REST and GraphQL API endpoints with structured assertions and reporting. Use when a user asks to test an API, hit an endpoint, check if an API works, validate a response, debug an API call, test authentication flows, or verify API contracts. Supports GET, POST, PUT, PATCH, DELETE with headers, body, auth, and response validation.
日本語の概要は準備中です。原文の説明を表示しています。
Manage public/private repository pairs. Verify parity, detect drift, run sync scripts, and validate no sensitive data leaks to the public repo. Use for projects that maintain separate dev and public repositories.
日本語の概要は準備中です。原文の説明を表示しています。
The systematic discipline of reviewing AI-generated and human-written marketing content for quality, accuracy, consistency, and effectiveness. This isn't proofreading—it's strategic quality control that catches issues before they damage brand, waste spend, or miss opportunities. In an era where AI can generate content at scale, QA becomes the bottleneck— and the moat. Speed without quality is just fast failure. Every piece of content is a brand promise, and QA ensures that promise is kept. Content QA operates at multiple levels: tactical (spelling, grammar), strategic (message alignment), technical (platform compliance), and brand (voice, tone). The best QA catches what the creator couldn't see because they were too close. Use when "review content, QA, quality check, proofread, check copy, content review, brand compliance, verify content, fact check, review before publish, content approval, final review, qa, content-review, quality-assurance, brand-compliance, proofreading, content-quality, review" mentioned.
日本語の概要は準備中です。原文の説明を表示しています。
Verify meeting notes are fully processed. Args: <path>, [dates], all, pending [dates] [--step=X]. Shows which wrap steps are complete or missing.
日本語の概要は準備中です。原文の説明を表示しています。
Audit experiment integrity against the evidence ledger. At L2 (repo + result files present) a fresh cross-model reviewer reads the eval code line-by-line for fake/derived ground truth, score self-normalization, phantom results (a paper number with no backing file/key), dead/uncalled metric code, verified-scope inflation, method-described ≠ method-evaluated drift, synthesized-looking results, placeholder/fake data still wired into a released result, code-output ≠ reported-number mismatch, and missing reproducibility artifacts (an empirical/agent/LLM paper shipping neither code nor the prompts/configs its results need) — every finding span-anchored to a ledger claim_id. At L0/L1 (PDF / source only) the same patterns are surfaced as info-level 'could-not-verify' signals where the ledger gives an anchor (observability_level_required:2) — NEVER a fraud verdict from a PDF. The reviewer PROPOSES findings; tools/adjudicate_findings.py computes the verdict. Detect-only. Triggers: "experiment forensics", "audit the results", "check the eval code", "实验诚实度".
日本語の概要は準備中です。原文の説明を表示しています。
Citation-integrity forensics: is every reference real, correctly attributed, and used in a context the cited work actually supports? Catches hallucinated references (no paper at the claimed arXiv id/DOI/venue, fabricated authors/year), metadata drift (wrong year/venue/version), and wrong-context citations (a real paper cited for a claim it never makes — or argues against). A hot zone for machine-generated papers. Decidable at L0 (text + canonical sources). Span-anchored to the evidence ledger (claims.json); the executor gathers canonical facts (DBLP / arXiv / DOI), then one FRESH cross-model thread per cited key proposes findings; reviewer != adjudicator. Emits citation-forensics.findings.json; NEVER computes the verdict. Triggers: "citation forensics", "check the references", "hallucinated citations", "wrong-context citation", "verify references", "引用核对".
日本語の概要は準備中です。原文の説明を表示しています。
Influencer Discovery workflow skill. Use this skill when the user needs Find and evaluate influencers for brand partnerships, verify authenticity, and track collaboration performance across Instagram, Facebook, YouTube, and TikTok and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.
日本語の概要は準備中です。原文の説明を表示しています。
Influencer Discovery workflow skill. Use this skill when the user needs Find and evaluate influencers for brand partnerships, verify authenticity, and track collaboration performance across Instagram, Facebook, YouTube, and TikTok and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.
日本語の概要は準備中です。原文の説明を表示しています。
表单保存后验证（Toast / 重定向 / 数据对比 / 富文本清空）。触发：保存验证、toast、重定向检测、Meta+A、保存成功判断。
日本語の概要は準備中です。原文の説明を表示しています。