Explore skills
Showing 46651–46680 of 77079 skills
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".
wanshuiyin/Anti-Autoresearch
Stars: 180Claude CodeCodexTatsh/patreon-archiver
Stars: 41Claude CodeCodexHeath's BRIEF prompting framework plus its eval harness. Use to WRITE a structured prompt for any go-to-market use case (SDR research, AE deal reads, CSM renewal calls, marketing lead scoring, RevOps analysis, win/loss, outreach, enrichment columns) and to MEASURE whether that prompt actually works by running an eval that reports a pass rate. BRIEF = Brief the goal, Role, Inputs, Execution, Finish line. Trigger on "write a prompt for", "build a BRIEF", "structure this prompt", "turn this into a prompt", "make this prompt better", "eval this prompt", "test this prompt", "what's the success rate of this prompt", "score this prompt", "is this prompt reliable", "A/B this prompt", or any request to author or evaluate a prompt, especially a repeatable one meant to run across many rows, deals, or accounts. Also fire when someone references SPICE and wants something more complete, or wants a prompt that produces the same quality every run.
heath-gtm/Skill-Builder
Stars: 0Claude CodeCodexThe mandatory structure for any leadership-facing strategic plan. Two bets max. Named team. Reads without scrolling. Use whenever the user asks for "a plan," "a 90-day plan," "a strategic plan," "a board update with a plan," "a leadership doc," "what we're going to do about X" — or anything that will be presented to a CEO, board, or exec team as "here's the plan." Trigger on "write the plan," "structure my plan," "tighten my plan," "the CEO wants the plan," or any leadership-strategy document where the test is "does the reader understand the plan in 30 seconds." Also fire as the mandatory structure for any pasted strategic doc being tightened by /heath-no-fluff or /heath-voice-humanizer.
heath-gtm/Skill-Builder
Stars: 0Claude CodeCodex以中文、English 或中英双语记录、解释和复盘成人日常排便与肠道健康信息,包括 Bristol 便形文字分类、频次和时段趋势、食物线索、可持续饮食调整、季节菜谱、每日排便提醒,以及危险信号分流和经同意的附近就医机构查找。Use for Chinese, English, or bilingual requests about poop, stool, bowel movements, constipation or diarrhea tendencies, foods associated with a bowel change, gut-friendly meals, recurring bowel-health reminders, or nearby appropriate care after a safety concern. Do not use as a diagnostic or treatment tool.
BulletsHo/la-le-me
Stars: 2Claude CodeCodexClarify ambiguous features, UX/UI changes, architecture decisions before coding.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
BumpyClock/agent-templates
Stars: 2Claude CodeCodexCreate distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexTeaches the AI to design like a high-end agency. Defines the exact fonts, spacing, shadows, card structures, and animations that make a website feel expensive. Blocks all the common defaults that make AI designs look cheap or generic.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexElite mobile app image-generation skill for creating premium, app-native screen concepts and flows. Designed for iOS, Android, and cross-platform mobile products. Prioritizes clean hierarchy, comfortably readable text, strong multi-screen consistency, controlled color palettes, non-generic creative direction, textured surfaces, image-led composition, tasteful custom iconography, and clean phone mockup framing. By default, screens should be shown inside a subtle premium iPhone or similar phone mockup with a visible frame, while the main focus stays on the app content itself. This skill generates images only. It does not write code.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexElite frontend image-direction skill for generating premium, conversion-aware website design references. CRITICAL OUTPUT RULE — generate ONE separate horizontal image FOR EVERY section. A landing page with 8 sections produces 8 images. Never compress multiple sections into one image. Enforces composition variety (not always left-text / right-image), background-image freedom, varied CTAs, varied hero scales (giant / mid / mini minimalist), narrative concept spine, second-read moments, and a single consistent palette across all images. Optimized for landing pages, marketing sites, and product comps that developers or coding models can accurately recreate.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexCode quality review for WinUI 3 apps — MVVM compliance, x:Bind correctness, accessibility, theming, security, and performance. Use before committing to catch issues that the compiler and UI tests won't find.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexCreate a durable, self-contained HTML canvas for standalone analyses, structured findings, and data exploration. Use when the artifact itself is the deliverable, not for incidental debugging output or work requested in another tool.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexVet a perf measurement (limiter, tuning, limits, errors, repeatability, relevance, and whether the work happened) before you report or act on it. Use when you run a benchmark or report a speedup or regression you measured.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when two or more fixes that share one premise have failed the same gate. Write the premise down and test it directly before you write another fix that assumes it.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when wiring validation, error handling, or framework adapters. Concentrate guards at system boundaries (CLI, config, network, external APIs); trust internal types and keep business logic in pure functions.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply to any non-trivial work, not just bulk work: edits, migrations, analyses, checks. Build the tool that does it or proves it (codemod, script, generator, or a skill your subagents follow) instead of working by hand. The tool is the artifact a reviewer can rerun.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexAssess delegation when work could use separate contexts, parallel execution, or a less capable model.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply before you measure, and before you trust, report, or act on a number you measured: a speedup, a regression, a throughput, a latency, or an eval result. Find what limits it, and rule out that it measured something other than the work you think.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when debugging. Trace each symptom back to the first place the system went wrong and fix it there. A guard is a fix only when the state it handles is legitimate.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when designing commands, lifecycle steps, or processing loops that run amid crashes, restarts, and retries. Converge to the same end state regardless of partial prior runs.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when reviewing or shaping code that's hard to trace. Count layers between question and answer, and hidden state in the reader's head; collapse one-caller wrappers and shrink mutable scope.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when writing stateful logic, or when code branches a lot or repeats a shape assumption across files.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when concurrent actors might write to the same file, branch, key, or state object. Eliminate the sharing first; serialize structurally only when one shared writer is a real invariant.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply to multi-step work (sweeps, migrations, runs of similar edits) and to how you stack commits and PRs. Break work into small units that each end in a verifiable state, check each before the next, and order delivery so the sequence proves itself to a reviewer.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when you write, change, or keep a test. Assert what the code's users can observe, so that the test fails when that behavior breaks and keeps passing when only the internals change.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexApply when designing types, reviewing a function signature, or writing code in any statically-typed language. Make illegal states unrepresentable, brand semantic primitives, parse external data at boundaries, exhaust variants, derive from authoritative schemas.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexRun an extremely strict maintainability review for abstraction quality, giant files, and spaghetti-condition growth. Use for a thermo-nuclear code quality review, thermonuclear review, deep code quality audit, or especially harsh maintainability review.
BumpyClock/agent-templates
Stars: 2Claude CodeCodexReview code changes using Hyperlane monorepo coding standards. Use when reviewing PRs, checking your own changes, or doing self-review before committing.
recoil-labs/recoil-link
Stars: 0Claude CodeCodexThe alphabetical-sort invariants the hyperlane-registry CI / CodeRabbit enforces on warp route YAML (deploy.yaml and config.yaml) — top-level chain entries sorted by chain name, and keys within each entry in strict alphabetical order — plus the canonical per-file key orderings and a verification pass. Referenced by any warp deploy/update skill that writes or edits a registry YAML file.
recoil-labs/recoil-link
Stars: 0Claude CodeCodex