Results for “pr”

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

Build or refactor CLI tools designed for AI agent consumption: non-interactive, flag-driven, idempotent, with --json output and --dry-run preview. Use when creating a new script the agent will call, adding agent-friendly flags to an existing tool, or debugging why an agent keeps failing to use your CLI.

Use this skill when creating a new Claude skill from scratch, editing or improving an existing skill, or measuring skill performance with evals and benchmarks. Invoke whenever the user says things like 'make a skill for X', 'turn this workflow into a skill', 'test my skill', 'improve my skill', 'run evals', 'benchmark this', or 'optimize my skill description'. Also use proactively when the conversation has produced a repeatable workflow that would benefit from being captured as a skill. Covers the full lifecycle: capture intent, draft SKILL.md, run evals, review with user, iterate, optimize description, package. NOT for general coding help, debugging runtime errors, building MCP servers, writing Claude hooks, or creating plugins - use domain-specific skills for those.

Statistical fine-mapping of GWAS loci using SuSiE, SuSiE-inf, and Approximate Bayes Factors to identify credible sets and posterior inclusion probabilities (PIPs) for causal variant discovery. SuSiE-inf adds an infinitesimal polygenic component for improved calibration at well-powered loci.

Deterministic mathematical computation using SymPy. Use for ANY math operation requiring exact/verified results - basic arithmetic, algebra (simplify, expand, factor, solve equations), calculus (derivatives, integrals, limits, series), linear algebra (matrices, determinants, eigenvalues), trigonometry, number theory (primes, GCD/LCM, factorization), and statistics. Ensures mathematical accuracy by using symbolic computation rather than LLM estimation.

Routes a research topic through a structured two-phase workflow: generate an extensible outline, then fan out parallel web-search agents to investigate each item into validated JSON, producing a complete markdown report with table of contents.

This skill should be used when the user asks to "test for insecure direct object references," "find IDOR vulnerabilities," "exploit broken access control," "enumerate user IDs or object references," or "bypass authorization to access other users' data." It provides comprehensive guidance for detecting, exploiting, and remediating IDOR vulnerabilities in web applications.

Rename Pharo methods safely through Genie's `rename_method` MCP refactoring tool. Use when a selector should be renamed while preserving behavior, updating senders and symbol references across the live image, including as one step in larger refactorings such as clarify API names, split responsibilities, or move behavior behind better messages.

Extend or continue an existing video clip on RunComfy via the `runcomfy` CLI. Routes to Google Veo 3-1's `extend-video` and `fast/extend-video` endpoints — pick the source video plus a prompt describing what should happen next, and the model produces a clip that continues the original with consistent motion, lighting, and subject identity. Use when the user has a short Veo clip and wants it longer, or wants a chained narrative built shot-by-shot from a single seed clip. Triggers on "extend video", "continue video", "longer video", "video extend", "make this clip longer", "Veo extend", "chain video shots", "video continuation", or any explicit ask to take an existing video and add more frames after it.

Extend or continue an existing video clip on RunComfy via the `runcomfy` CLI. Routes to Google Veo 3-1's `extend-video` and `fast/extend-video` endpoints — pick the source video plus a prompt describing what should happen next, and the model produces a clip that continues the original with consistent motion, lighting, and subject identity. Use when the user has a short Veo clip and wants it longer, or wants a chained narrative built shot-by-shot from a single seed clip. Triggers on "extend video", "continue video", "longer video", "video extend", "make this clip longer", "Veo extend", "chain video shots", "video continuation", or any explicit ask to take an existing video and add more frames after it.

Guides usage of the Gemini API on Agent Platform with the Google Gen AI SDK, covering SDK usage (Python, JS/TS, Go, Java, C#), capabilities like multimodal inputs, tools, media generation, caching, batch prediction, and Live API.

Deploy and operate the VSS 3.2 GA RT-Embed Video Embedding microservice using Docker Compose, covering GPU prerequisites, REST API usage for file uploads, text/video embeddings, live RTSP streams, Redis/Kafka/OTel integration, and troubleshooting.

Package an in-progress task on this cluster so a Claude agent on another cluster (MMM Young, CX3, HX1, Isambard) can pick it up, or resume a task handed off to this cluster. Use when the user says hand off / migrate / continue this on <cluster>, or asks you to pick up an existing handoff.

Classifies data as pseudonymised or anonymised using Recital 26 reasonably likely test, Breyer ruling C-582/14, motivated intruder test, and WP29 Opinion 05/2014 on anonymisation techniques. Covers singling out, linkability, and inference tests. Keywords: pseudonymisation, anonymisation, Recital 26, re-identification, k-anonymity, differential privacy, WP29 Opinion 05/2014.

Scaffold and implement Model Context Protocol (MCP) servers that expose external services, APIs, and data sources as typed tools and resources for LLM agents. Use when the user says "build an MCP server", "give Claude access to X", "create an MCP tool", "expose my API to an agent", or "AI agent integration".

Orchestrate a configurable, multi-member CLI planning council (Codex, Claude Code, Gemini, OpenCode, or custom) to produce independent implementation plans, anonymize and randomize them, then judge and merge into one final plan. Use when you need a robust, bias-resistant planning workflow, structured JSON outputs, retries, and failure handling across multiple CLI agents.

Read and search GitHub repository documentation via gitmcp.io MCP service. **WHEN TO USE:** - User provides a GitHub URL - User mentions a specific repo in owner/repo format - User asks "what does this repo do?", "read the docs for X repo", or similar - User wants to search code or docs within a repo

Provides 25 paid HTTP endpoints on Base mainnet that bill agents per call in USDC for web search, AI analysis, crypto and stock data, SEC filings, company intelligence, news, sentiment scoring, and a macro dashboard, returning structured JSON.

Multi-agent board meeting protocol for strategic decisions. Runs a structured 6-phase deliberation: context loading, independent C-suite contributions (isolated, no cross-pollination), critic analysis, synthesis, founder review, and decision extraction. Use when the user invokes /cs:board, calls a board meeting, or wants structured multi-perspective executive deliberation on a strategic question.

Use for writing, reviewing, debugging, testing, or optimizing Python CVXPY optimization models. Trigger on Variable, Parameter, Expression, Constraint, Objective, Problem, DCP, DPP, DGP, DQCP, solver selection/status, dual values, mixed-integer, cone, or repeated parametric solves. Do not use for scipy.optimize-only, PyMC inference, symbolic algebra without optimization, or hand-written solver implementations.

Meta-agent for creating new custom agents, skills, and MCP integrations. Expert in agent design, MCP development, skill architecture, and rapid prototyping. Activate on 'create agent', 'new skill', 'MCP server', 'custom tool', 'agent design'. NOT for using existing agents (invoke them directly), general coding (use language-specific skills), or infrastructure setup (use deployment-engineer).

Design, create, audit, and improve Claude Agent Skills with expert-level progressive disclosure. Use when building new skills, reviewing existing skills, debugging activation failures, encoding domain expertise, designing skills for subagent consumption, or understanding platform constraints and distribution surfaces. NOT for general Claude Code features, runtime debugging, non-skill coding, or MCP server implementation.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.

自动捕获经验并安全进化的技能。触发条件：(1)命令/操作失败时→记ERRORS.md (2)被用户纠正('不对'/'应该是')时→记LEARNINGS.md (3)用户需要不存在的能力时→记FEATURE_REQUESTS.md (4)外部API/工具出错时→记ERRORS.md (5)发现自己知识过时/错误时→记LEARNINGS.md (6)发现更好做法时→记LEARNINGS.md (7)每个任务完成时→回顾过程，有新经验则记LEARNINGS.md。去重原则：如果没有新经验或已有条目已覆盖则跳过不写。每次写入同时在.learnings/CHANGELOG.md追加JSONL日志。经验反复出现≥3次时晋升到AGENTS.md/TOOLS.md/SOUL.md。详见正文。

Design and verify auditable human oversight, approval gates, escalation paths, and safe state transitions for AI agent workflows. Use when deciding which agent actions require review, adding approve/reject or dual-control flows, preventing unauthorized autonomous effects, creating decision records, reducing rubber-stamping, or recovering safely from rejected, expired, or failed actions.

Control tmux panes and communicate between AI agents. Use this skill whenever the user mentions tmux panes, cross-pane communication, sending messages to other agents, reading other panes, managing tmux sessions, or interacting with processes running in tmux. Includes tmux-bridge CLI for agent-to-agent messaging and raw tmux commands for direct session control.

Generates a 15-second cinematic cooking tutorial video from a single person photo, using a two-step process that first creates a composite reference sheet and then animates the full sequence with consistent character and environment.

Run a reproduce → isolate → verify debugging workflow for concrete bugs, regressions, flaky failures, and environment-specific behavior. Use when the user already has a failing command, test, request, UI flow, or narrowed symptom and needs root-cause diagnosis or fix verification rather than raw log-line selection, broad test-policy design, PR review, or generic performance tuning.

This skill should be used when working with reinforcement learning tasks including high-performance RL training, custom environment development, vectorized parallel simulation, multi-agent systems, or integration with existing RL environments (Gymnasium, PettingZoo, Atari, Procgen, etc.). Use this skill for implementing PPO training, creating PufferEnv environments, optimizing RL performance, or developing policies with CNNs/LSTMs.

Review Markdown prose against agent-style's 21 rules. `/style-review FILE` audits the file and (on user confirm) writes a polished copy at `FILE.reviewed.md`. `/style-review A.md B.md` A/B-compares two drafts and emits a per-rule delta table. Complements the generation-time rules installed via `agent-style enable <tool>`.

Build multi-agent AI systems with LangGraph — the low-level orchestration framework for stateful, graph-based agent workflows. Covers supervisor, swarm, and hierarchical multi-agent patterns; subgraph composition; state management (checkpointers/stores); persistence; evals; and production debugging. Reach for this when designing agent architectures that need cycles, conditional branching, parallel execution, or human-in-the-loop patterns.

Compact the current conversation into a handoff document for another agent to pick up. Save to a user-configured location (OS temp, home folder, or per-project .handoff/), redact secrets before write, suggest skills for the next session, and auto-load the latest handoff on the next SessionStart. First-run setup asks where to save so the project folder never gets cluttered. Use when the user says 'hand this off', 'handoff doc', 'summarize this for a new session', 'compact this conversation', 'I'm ending this session', 'pick this up later', or any variation signaling intent to pass work to a fresh agent. Also trigger on implicit signals: the user announcing they're switching machines, ending the day mid-task, or context is growing long without a natural stopping point.

Deep API security assessment beyond surface scanning. Covers the full OWASP API Security Top 10 (2023): Broken Object Level Authorization (BOLA / IDOR), Broken Authentication, Broken Object Property Level Authorization (mass assignment + excessive data exposure), Unrestricted Resource Consumption, Broken Function Level Authorization (BFLA / vertical privilege escalation), Unrestricted Access to Sensitive Business Flows, Server-Side Request Forgery via API parameters, Security Misconfiguration, Improper Inventory Management (shadow/zombie/deprecated endpoints, v1/v2 drift), and Unsafe Consumption of third-party APIs. Works across REST, GraphQL, gRPC, SOAP, and MCP servers. Discovers APIs from OpenAPI/Swagger specs, GraphQL introspection, gRPC reflection, .well-known endpoints, JS bundles, and traffic capture. Uses kiterunner, ffuf, schemathesis, restler-fuzzer, openapi-fuzzer, graphql-cop, clairvoyance, batchql, inql, jwt_tool, postman, mitmproxy, and manual http(action="request", ...) payloads. Every techniqu

Onboard an autonomous agent to the tiny.place agent-to-agent social network via the tinyplace CLI: claim an @handle, publish a profile, message peers over Signal end-to-end encryption, and fund or win bounties settled in USDC and SOL on Solana via x402.

Compiles any research input — PDF papers, GitHub repositories, experiment logs, code directories, or raw notes — into a complete Agent-Native Research Artifact (ARA) with cognitive layer (claims, concepts, heuristics), physical layer (configs, code stubs), exploration graph, and grounded evidence. Use when ingesting a.