发现 Skills

Zero-friction idea capture. Append, list, or promote notes to todos.

When the user wants to create SEO-driven pages at scale using templates and data. Also use when the user mentions "programmatic SEO," "template pages," "pages at scale," "directory pages," "location pages," "[keyword] + [city] pages," "comparison pages," "integration pages," or "building many pages for SEO." For auditing existing SEO issues, see seo-audit.

知识星球用户信息：查看当前登录用户的个人资料、查询跨星球的最近发主题足迹。当用户需要查看自己的用户 ID、昵称、头像、认证状态，或查看自己最近在各星球发过的主题时使用。

Structured multi-agent deliberation for complex, multi-tradeoff problems — architecture design, system proposals, high-stakes decisions. Spawns competing-perspective agents that research and argue from constrained value functions. Trigger: "run a council", "council this", "explore from multiple angles", "design a council for". Heavyweight — do NOT trigger for simple questions or problems with one obvious direction.

Add a real-time token-usage footer to every Hermes response in CLI and Gateway by modifying the agent core and display layers.

Produce a complete firmware architecture spec for a described device — layer diagram, module responsibilities, HAL interface definitions, key state machines, RTOS decision. Use when asked to "design firmware architecture", "plan embedded firmware", "architect an IoT device", "how should I structure this firmware", or given a device description and asked what the firmware should look like.

Evaluates any proposed trading strategy as a complete mechanical engine — stress-testing for component completeness, logical alignment, mathematical capacity, and structural efficiency. Takes a strategy_manifest (philosophy, setup, risk, tactics, routine) and user_thresholds (target CAGR, max drawdown, max consecutive losses) as inputs. Runs 4 steps: Completeness Audit, Alignment & Bonds Test, Monte Carlo Overheating Simulation (1000 iterations), and Efficiency Re-routing. Returns an audit_report with is_deployable boolean, missing components, alignment errors, capacity warnings, and optimization guidance.

This skill should be used when users ask to "create a skill", "validate a skill for quality", "evaluate skill improvements", "improve my skill", "update my skill", "fix skill", "iterate on skill", "optimize skill", "skill quality", "skill performance", "skill isn't working", "analyze skill metrics", "init a new skill", "check skill compliance", or "sync skill to marketplace". Provides comprehensive skill development with automated validation, metrics tracking, and improvement workflows.

Meta's 86M prompt injection and jailbreak detector. Filters malicious prompts and third-party data for LLM apps. 99%+ TPR, <1% FPR. Fast (<2ms GPU). Multilingual (8 languages). Deploy with HuggingFace or batch processing for RAG security.

Open-source intelligence gathering - company repository enumeration, secret scanning, git history analysis, employee footprint, and code exposure discovery.

Autonomous multi-agent coding with git worktree isolation, QA validation, and memory. Use for complex features requiring autonomous implementation.

Volunteer management system design with scheduling, skill matching, communication templates, recognition programs, and retention strategies. Use when building volunteer programs, coordinating volunteers, or improving volunteer engagement.

Create guitar-style audio drafts from songs, melody descriptions, MIDI-derived notes, or hummed/sung melody references. Use when Codex should extract or accept a main melody, map it to a playable guitar register, and render a guitar audio file such as WAV; do not use when the requested deliverable is notation or a complete multi-part guitar arrangement.

Market selection, localization, operations, and regulatory constraints for global launches.

Cloud, DevOps, observability, and systems design guidance for scaling from thousands to millions of users.

Interact with a UniFi Dream Router (UDR) or any UniFi OS console via its REST API. Use this skill whenever the user wants to query or control their UniFi network — listing connected clients, blocking/unblocking devices, managing firewall policies, checking WAN health and speed test results, rebooting devices, managing VLANs or SSIDs, reading traffic stats, port forwarding, or any other UniFi network management task. Trigger even if the user doesn't say "API" or "UniFi" — phrases like "check my network", "block that device", "show me who's connected", "add a firewall rule", "what's my WAN IP", "how's my internet speed", or "what's on the guest network" are all good triggers.

TrussC framework knowledge — API patterns, node system, graphics, build workflow, common pitfalls. Use when writing TrussC C++ code.

Apply for pytorch-scikit-learn-cursorrules-prompt-file. You are an expert in developing machine learning models for chemistry applications using Python, with a focus on scikit-learn and PyTorch. Key Principles: - Write clear, technical responses with preci

Continue working on an OpenSpec change by creating the next artifact. Use when the user wants to progress their change, create the next artifact, or continue their workflow.

Load a customer's pricing guide before billing work. Use when working on billing rules, investigating discrepancies, or configuring charges for any customer. MUST be invoked before asking questions about rates, rules, or charge logic.

Define types, entities, and central constant registrations for syncable entities in Twenty's workspace migration system. Use when creating new syncable entities, defining TypeORM entities, flat entity types, or registering in central constants (ALL_ENTITY_PROPERTIES_CONFIGURATION_BY_METADATA_NAME, ALL_ONE_TO_MANY_METADATA_RELATIONS, ALL_MANY_TO_ONE_METADATA_FOREIGN_KEY, ALL_MANY_TO_ONE_METADATA_RELATIONS).

Send emails via system automation. Use when asked to "send email", "email someone", or when any agent needs to send an email. Supports single/bulk sending, templates, and attachments.

Persistent two-tier memory for AI agents. Use when storing learnings, searching past work, checking memory health, or following the session start/end memory protocol.

Static analysis and security scanning using Semgrep, a fast, open-source static analysis tool for finding bugs and enforcing code standards. Use when scanning code for security vulnerabilities, code quality issues, style violations, or bugs. Triggers include requests to find security issues, detect code patterns, enforce coding standards, scan for vulnerabilities, write custom rules, perform static analysis, or audit code quality. Supports pattern matching, autofix, generic pattern matching, and rule composition.