Agent skills you can check before you install
Public SKILL.md skills from GitHub repositories with 50+ stars and from the vendors themselves, scanned and scored. Install any of them with one command in Claude Code, Cursor, Codex, GitHub Copilot or Gemini CLI.
48,622 skills from 2,355 repositories · updated 2 hours ago
48,622 skills
- principle-attack-the-premisebacknotprop/pstackApply 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.TestingNo warnings★ 3.3K↓ —
- gsd:add-testsdavepoon/buildwithclaudeGenerate tests for a completed phase based on UAT criteria and implementationTestingNo warnings★ 3.7K↓ —
- principle-laziness-protocolbacknotprop/pstackApply when refactoring, evaluating diff size, or tempted to add abstractions, layers, or signal threading. Bias toward deletion and the smallest change that solves the problem.EngineeringNo warnings★ 3.3K↓ —Also in 1 other repo
- oil-motionnexu-io/open-design处理视频和序列帧，编译滚动、指针、拖动与状态驱动的网页动效；内含本地媒体处理脚本和运行时。新素材生成使用宿主工具。OtherNo warnings★ 105.3K↓ —
- debriefdavepoon/buildwithclaudeAlias for `/origin:handoff` — symmetric brief/debrief naming. Same behavior: end-of-session ritual that writes session log + project status + granular MCP captures. Invoked as `/debrief`. Use when the user prefers the brief/debrief pair over brief/handoff.WritingNo warnings★ 3.7K↓ —
- html-planplannotator/effective-htmlDirect-invocation specialist for clear, self-contained HTML plans that preserve source material while improving hierarchy, sequence, ownership, dependencies, and reviewability. Use when the user explicitly invokes html-plan or the broad html skill routes a plan request here. Do not activate independently from a general request.PlanningNo warnings★ 3.8K↓ —
- deliverability-qaaaron-he-zhu/aaron-marketing-skillsUse when the user asks to "run a deliverability pre-flight before I send", "check my SPF/DKIM/DMARC/BIMI", "why am I landing in spam / promotions", or "score my sender reputation and list hygiene"; runs the ONE-TIME pre-send SEND S1 authentication pre-flight and builds the SEND S (Sender-integrity / Deliverability) evidence read — DNS + DMARC-RUA auth, domain/IP reputation, inbox placement, content/link/render, and point-in-time bounce/complaint hygiene — using Pass/Partial/Fail/Unknown/N/A states and scoring only at complete applicable coverage. Not for the recurring hygiene trend — use list-hygiene-monitor; not for final EQS or veto verdicts — use email-quality-auditor; not for segments/suppression lists — use list-segment-builder. 邮件送达率预检/SPF DKIM DMARC认证/发件域声誉TestingNo warnings★ 3K↓ —
- principle-fix-root-causesbacknotprop/pstackApply 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.EngineeringNo warnings★ 3.3K↓ —Also in 1 other repo
- figma-implement-designforyourhealth111-pixel/Vibe-SkillsTranslate Figma nodes into production-ready code with 1:1 visual fidelity using the Figma MCP workflow (design context, screenshots, assets, and project-convention translation). Trigger when the user provides Figma URLs or node IDs, or asks to implement designs or components that must match Figma specs. Requires a working Figma MCP server connection.DesignNo warnings★ 3.8K↓ —
- typescript-best-practicesbacknotprop/pstackTypeScript best practices. Use when reading or editing any .ts or .tsx file.EngineeringNo warnings★ 3.3K↓ —
- ulw-researchcode-yeongyu/oh-my-openagentRuns maximum-saturation research with a cooperating team, claim-graph gating, and a cited, QA'd deliverable. Use when the user explicitly asks for research or a deep investigation, including any 'ulw' research wording.ScienceNo warnings★ 70.6K↓ —
- postwiredavepoon/buildwithclaudePublish and schedule to TikTok, Instagram, YouTube, LinkedIn, X, Bluesky, Mastodon, Facebook, Threads, Reddit, Telegram and Discord through one PostWire API call — writing a separate, native version of the post for each network instead of sending identical text everywhere. Use when the user wants to publish or schedule one idea across several platforms, wants per-network captions, or wants to know why a post was rejected before it is sent.MarketingNo warnings★ 3.7K↓ —
- tao-train-mask2formerNVIDIA/skillsMask2Former for universal image segmentation (panoptic, instance, and semantic). Transformer-based with masked attention for high-quality segmentation results. Use when training, evaluating, exporting, quantizing, or running inference for a TAO Mask2Former model. Trigger phrases include "train Mask2Former", "universal segmentation", "panoptic / instance / semantic segmentation", "masked-attention transformer segmenter".MediaNo warnings★ 3.8K↓ —
- subject-line-labaaron-he-zhu/aaron-marketing-skillsUse when the user asks to "generate subject line variants", "pre-score my subject lines", or "will this subject get truncated / trigger spam filters"; produces a labeled subject + preheader variant set and a per-variant heuristic pre-score card — spam-trigger flags, length/truncation across desktop + mobile, emoji-count, and the inbox preview render (from-name + subject + preheader) — before any test is run. Not for the body copy or CTA — use email-creative-builder; not for the A/B test design or significance read — use send-experiment-designer; not for the profile-weighted EQS or the S1/S2/N1/D1 vetoes — use email-quality-auditor. 邮件主题行生成/主题行预打分/截断与垃圾词检查TestingNo warnings★ 3K↓ —
- gsd:undodavepoon/buildwithclaudeSafe git revert. Roll back phase or plan commits using the phase manifest with dependency checks.PlanningNo warnings★ 3.7K↓ —
- gsd:nextdavepoon/buildwithclaudeAutomatically advance to the next logical step in the GSD workflowProductivityNo warnings★ 3.7K↓ —
- gsd:spikedavepoon/buildwithclaudeSpike an idea through experiential exploration, or propose what to spike next (frontier mode)OtherNo warnings★ 3.7K↓ —
- html-prototypeplannotator/effective-htmlDirect-invocation specialist for polished, responsive, self-contained HTML mockups and interactive prototypes grounded in the user's conversation, product context, and design language. Use when the user explicitly invokes html-prototype or the broad html skill routes a mockup or prototype request here. Do not activate independently from a general request. Treat a mockup as a noninteractive fidelity mode within this skill, not as a separate skill.DesignNo warnings★ 3.8K↓ —
- serp-analysisaaron-he-zhu/aaron-marketing-skillsUse when the user asks to "analyze the SERP" or "SERP分析"; maps SERP features, layout, ranking factors, search intent, AI Overviews, and snippet opportunities for a query. Not for keyword demand discovery — use keyword-research. SERP分析/搜索结果ScienceNo warnings★ 3K↓ —
- memory-managementaaron-he-zhu/aaron-marketing-skillsUse when the user asks to "remember project context", review saved findings, initialize runtime memory, archive stale work, reconcile notes, or erase a subject; manages authorized HOT/WARM/COLD working memory across all disciplines while preserving registry event ownership and privacy controls. Not for changing canonical registry facts - route those through the owning registry. 项目记忆/跨会话ProductivityNo warnings★ 3K↓ —
- technical-writingbacknotprop/pstackLayered technical-writing standard: Diátaxis structure, Google developer style sentences, STE instruction rules, Global English syntax. Use for /technical-writing or when writing or reviewing docs, RFCs, readmes, PR descriptions, or commit messages.WritingNo warnings★ 3.3K↓ —
- principle-build-the-leverbacknotprop/pstackApply 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.WritingNo warnings★ 3.3K↓ —
- opendavepoon/buildwithclaudeBoot the session, load context from the vault, and surface what mattersProductivityNo warnings★ 3.7K↓ —
- principle-exhaust-the-design-spacemichael-denyer/pstack-claudeApply when facing a novel UI interaction or architectural decision with no precedent in the codebase. Build 2-3 competing prototypes and compare side by side before committing.DesignNo warnings★ 3.7K↓ —
tao-finetune-clip
NVIDIA/skills
DevOpsUpdated 2 days ago
From the author
CLIP vision-language model for image-text retrieval, zero-shot classification, embedding extraction, ONNX export, and TensorRT deployment. Use when fine-tuning or training CLIP, running zero-shot classification, computing image embeddings, or deploying CLIP to ONNX/TensorRT. This is a single-action model skill; do not use it for an iterative weak-attribute improvement loop that keeps retraining and evaluating until progress stops, which belongs to tao-run-deft-pas.
Install
npx skills add NVIDIA/skills --skill tao-finetune-clip -a claude-code
Installs to .claude/skills/tao-finetune-clip/ · Each command fetches the skill straight from GitHub with the open skills CLI.
Download .zip
Byte-identical to the scanned commit, with the repository’s licence and a SOURCE.txt.
What it can touch
Instructions onlyNo network use seenLicence: APACHE-2.0
Safety checks on SKILL.md
No warnings.
Why 90
Security100
Provenance83
Maintenance100
Licence100
Adoption57
Files (23)
- BENCHMARK.md
- SKILL.md
- config/skillspector-baseline.yaml
- evals/evals.json
- references/clip-peft.md
- references/error-patterns.md
- references/pas-metrics.md
- references/skill_info.yaml
- references/spec-param-inference.md
- references/spec_template.yaml
- references/spec_template_deploy.yaml
- references/spec_template_evaluate.yaml
- +11 more