All categories
Documentation
Docs, README files, API references, tutorials, specs, and changelogs
- 16,089
- 671
Security grades appear on each card once the skill has been scanned. Newly imported skills may briefly show without a grade until the backfill job runs.
Open in full browserBrowse documentation skills
Showing 6,889–6,912 of 16,089 skills
- Skill Improve使用测试-修复-重测循环改进技能，运行 skill-test 识别问题，提出修复方案，用户批准后重写 SKILL.md 并重测验证。Votes: 0GitHub stars: 2
- Prototype技术原型验证，验证关键技术点可行性，产出 PROCEEDS 或 PIVOTS 结论。Votes: 0GitHub stars: 2
- Dev Story核心实现技能。读取 Story 文件并按 7 阶段工作流实现功能：找到 Story、加载上下文、路由到正确程序员、实现、测试证据、收集总结、更新会话状态。Votes: 0GitHub stars: 2
- Create Skill创建新技能，通过提案-草稿-审核-批准-注册-同步六阶段工作流，确保新技能定义符合 Agent Skills 开放标准和项目模式规范。Votes: 0GitHub stars: 2
- Create Rule创建新路径规则，通过提案-草稿-审核-批准-注册-同步六阶段工作流，确保新规则符合路径规则模式规范并不与现有规则冲突。Votes: 0GitHub stars: 2
- Tech Spec> **Source**: `../aid-tech-spec/SKILL.md` > > This file is a reference. The main skill definition is in the `.claude/skills/` directory. > Use the `/tech-spec` command for technical specification generation.Votes: 0GitHub stars: 9
- Cucumber BddCucumber BDD integration for AID methodology. Gherkin-first acceptance criteria with executable specifications. Use for writing feature files, step definitions, and running acceptance tests.Votes: 0GitHub stars: 9
- Aid PrdAID Phase 1 - PRD creation. Use for user stories, acceptance criteria, scoping features, transitioning from discovery to tech spec.Votes: 0GitHub stars: 9
- Aid DevelopmentAID Phase 4 - Development phase. Use for implementing features, TDD practices, code reviews, transitioning from planning to QA.Votes: 0GitHub stars: 9
- Abot Physworld Physics Aligned World ModelReplace standard likelihood-based video diffusion training with decoupled physics discriminators and DPO post-training to suppress physically implausible behaviors (object penetration, anti-gravity motion) in robotic manipulation videos. Use when generating physics-realistic video predictions for embodied AI and want to maintain visual quality without physical violations.Votes: 0GitHub stars: 6
- Wan Move Motion ControlEnable precise motion control in video generation using dense point trajectories as latent space features. Wan-Move integrates with existing models without architecture changes—ideal when you need fine-grained scene control without auxiliary motion encoders.Votes: 0GitHub stars: 6
- Voyager 3d Scene GenerationGenerate spatially-coherent 3D point-cloud videos from single images using depth-fused diffusion with efficient world caching for infinite scene exploration.Votes: 0GitHub stars: 6
- Vla Rft World Model Reinforcement RoboticsFine-tune Vision-Language-Action models using learned world models as simulators, eliminating costly real-world or physics-simulation RL. Train robust robot policies in 400 steps via GRPO with model-generated verified rewards.Votes: 0GitHub stars: 6
- Videovla Robot ManipulatorsTransform video generation models into robot manipulators by jointly predicting actions and future visual outcomes. VideoVLA demonstrates that imagining futures improves action reliability—ideal when you need generalizable robot learning from video imagination.Votes: 0GitHub stars: 6
- Vespo Variational Sequence RlStabilize off-policy RL training for LLMs by deriving principled importance weight reshaping from variational optimization. Instead of heuristic clamping, VESPO uses closed-form exponential weighting W^c1 * exp(c2*(1-W)) to suppress overweighted samples while maintaining smooth gradients. Enables stable training at 64× policy staleness and under fully asynchronous execution with sequence-level operations that avoid length-dependent biases.Votes: 0GitHub stars: 6
- Trapo Semi Supervised RlBridge labeled and unlabeled data through trajectory similarity in reinforcement learning. Select reliable unlabeled samples by comparing pass-rate evolution trajectories against labeled data. Achieve 42.6% accuracy with 1K labeled + 3K unlabeled samples, surpassing fully-supervised training on 45K labels with 10% annotation budget.Votes: 0GitHub stars: 6
- Trajectory Selection ReasoningSelect best reasoning trajectories from multiple samples using step-level scoring from a 0.6B lightweight verifier that exploits hidden states, outperforming external reward models by 4-12% without massive annotations.Votes: 0GitHub stars: 6
- Test Time Rl VlmAdapt vision-language models at inference without labeled data by generating multiple predictions and rewarding high-frequency outputs. Trigger: improve VLM accuracy on deployment with self-generated supervision signals.Votes: 0GitHub stars: 6
- Svg T2i Vfm DiffusionTrain text-to-image diffusion models directly in frozen DINOv3 feature spaces, eliminating VAE-based compression. Enables high-resolution synthesis by leveraging VFM representations as native latent manifolds with unified cross-modal transformers.Votes: 0GitHub stars: 6
- Surgical Post Training Error CorrectionCorrect reasoning errors with minimal data collection by using an oracle to surgically fix only erroneous steps in existing trajectories. Use binary classification loss on rectified pairs with implicit KL regularization to prevent knowledge forgetting.Votes: 0GitHub stars: 6
- Sr Grpo Stable RankUses stable rank (variance distribution across semantic dimensions) as annotation-free reward signal in GRPO to incentivize high-dimensional representation structures, eliminating dependency on human preference annotations or learned reward models.Votes: 0GitHub stars: 6
- Spatial Ssrl Spatial Understanding RlImprove spatial reasoning in vision-language models through five automatically-formulated pretext tasks (patch shuffling, flipping, inpainting, depth ordering, 3D position) that require zero human annotation, using only RGB/RGB-D images and RL optimization for 4% accuracy gains.Votes: 0GitHub stars: 6
- Sage Long Video AgentsEnable agents to reason over long videos through multi-turn reasoning for complex questions and single-turn answering for simpler queries. Equip orchestrator VLM with multiple tools (web search, speech transcription, temporal grounding). Use synthetic data generation and multi-reward RL achieving 6.1% improvements on open-ended tasks.Votes: 0GitHub stars: 6
- Repair Robust Editing LlmsPerform robust, low-cost model updates via REPAIR: closed-loop feedback, dynamic memory management, and frequent knowledge fusion. Preserve non-target knowledge while achieving 10-30% improvements in editing accuracy. Enable sequential edits without catastrophic forgetting through strong locality guards and stable integration mechanisms.Votes: 0GitHub stars: 6