发现 Skills

Generate a custom checklist for the current feature based on user requirements.

Create a feature branch with sequential or timestamp numbering

A Skill for converting DeepFRI from TensorFlow to PyTorch and migrating it to Ascend NPU. Suitable for TF model analysis, PyTorch reimplementation, layer-by-layer weight mapping, NPU inference and accuracy verification in protein function prediction scenarios—especially when you need to run DeepFRI’s CNN or GCN pathways on Ascend.

MuXi dense and sparse linear solver library that provides direct solvers (LU, Cholesky, QR, SVD) and eigensolvers. Intended for scientific computing, computer vision, linear optimization and related fields.

Automated code review with focus on security, performance, and maintainability. Use when reviewing code changes, PRs, or performing scheduled audits.

Analyzes the current project structure and tech stack. Use when asked to explore, understand, or summarize a project. Trigger terms: project overview, analyze codebase, what is this project.

Refactor MoonBit code to be idiomatic: shrink public APIs, convert functions to methods, use pattern matching with views, add loop invariants, and ensure test coverage without regressions. Use when updating MoonBit packages or refactoring MoonBit APIs, modules, or tests.

Access the RCSB Protein Data Bank (PDB) for 3D protein and nucleic acid structures. Search by text, sequence, or structure; download coordinate files (PDB/mmCIF/BinaryCIF); and retrieve rich metadata for structural biology and drug-discovery workflows.

Create executive presentations using StarHub's official templates and formatting standards. This skill should be used when building Board updates, CLT presentations, BPR decks, or any executive-level PowerPoint for StarHub. Includes the official CX Board template with proper slide layouts and styling.

Write a product requirements document (PRD) that captures problem, goal, scope, approach, and risk. Use whenever the user mentions PRD, product requirements, product doc, product brief, requirements doc, or needs to write a structured document about what to build and why.

Search Xiaohongshu (小红书) for trending content, product reviews, creator analytics, and lifestyle recommendations. Use when user asks about RedNote, 小红书, product reviews, Chinese social commerce, trending topics on RedNote, or wants to find popular posts and creators.

Manage project dependencies effectively. Use when adding, updating, or auditing dependencies. Covers version management, security scanning, and lockfiles.

Identify and analyze security threats. Use when designing systems, reviewing architecture, or assessing risk. Covers STRIDE methodology.

Walk a BloodHound Enterprise deployment's active attack-path findings, prioritise them by severity and exposure, and produce a ranked action list. Use when the caller asks "what should we fix first", "what attack paths are active", "review BHE findings", or after an ingest cycle that surfaced new findings.

Burp Suite Professional MCP integration reference. Use when working with Burp proxy history, sending requests through Burp, using Repeater/Intruder, checking scanner issues, or performing OOB testing with Collaborator.

Sentinel-V exploit verification agent for confirming bugs and exploits across web application, API, and AI/LLM pentesting. Use when verifying a reported vulnerability, confirming a finding, proving/disproving an exploit, generating a verification report, or checking for false positives.

Structured first-pass triage of an unknown memory image. Use when handed a .mem/.raw/.vmem/.dmp/.lime/.bin/.aff4 and asked "what happened on this box?" — establishes ground truth before any targeted hunt.

Upgrade blind SSRF to visible using HTTP redirect loops and error differential analysis. Use when blind SSRF confirmed but response content not observable.

This skill is used by **Goalkeeper** to manage and compress the task context, preventing token bloat in long-running tasks.

Analyze notes for quality, accuracy, and completeness. Use when asked to validate, review, check, update, expand, or improve a personal knowledge base note.

Use when a codebase feels hard to change, test, or navigate — surface architectural friction, identify deeper module seams, and walk one candidate into a concrete refactoring direction.

Use when a request arrives and the right execution path is unclear — routes the work to the correct mode, agent type, model tier, and delegation pattern before implementation starts.

Combined backend protocols for authentication (Clerk JWT verification) and PII-safe observability. Use when handling auth on the server, verifying Clerk webhooks via `svix`, scoping metadata via `publicMetadata`/ `privateMetadata`, or sanitizing logs to keep emails, tokens, and request bodies out of telemetry.

Simplifies code for clarity. Use when refactoring code for clarity without changing behavior. Use when code works but is harder to read, maintain, or extend than it should be. Use when reviewing code that has accumulated unnecessary complexity.