发现 Skills

Implement particle-object collisions for interactive particle effects

Query ChEMBL's bioactive molecules and drug discovery data. Search compounds by structure/properties, retrieve bioactivity data (IC50, Ki), find inhibitors, perform SAR studies, for medicinal chemistry.

Content caster (铸). Converts content into PNG visuals. Six molds: -l (default) long reading card, -i infographic, -m multi-card reading cards (1080x1440), -v visual sketchnote, -c comic (manga-style B&W), -w whiteboard (marker-style board layout). Exports to ~/Downloads/. Triggered by user commands such as 'cast', 'make image', 'make cards', 'make infographic', 'make poster', 'visual note', 'sketchnote', 'comic', 'manga', 'whiteboard'. Replaces ljg-cards and ljg-infograph.

Archives WeChat Official Account articles to a knowledge base and generates auto-created structured notes. Use when users provide WeChat article URLs (mp.weixin.qq.com) or mention archiving/saving WeChat articles. Combines the wechat2md scraper and note-creator for a complete workflow.

Write idiomatic C++ code with modern features, RAII, smart pointers, and STL algorithms. Handles templates, move semantics, and performance optimization.

Create a Positioning & Messaging Pack (positioning statement, messaging hierarchy, value proposition, tagline/headlines, copy set, validation plan). Use for positioning, messaging, value prop, tagline, homepage hero, one-liner, elevator pitch, and press pattern-matching. Category: Marketing.

Run blameless post-mortems & retrospectives and produce a Post-mortems & Retrospectives Pack (brief + agenda, facts/timeline, contributing factors + root causes, decisions + action tracker, kill criteria, learning dissemination plan). Use for postmortem, post-mortem, retrospective, retro, after action review, lessons learned. Category: Leadership.

Reverse engineers binaries using Ghidra's headless analyzer. Use when decompiling executables, extracting functions, strings, symbols, or analyzing call graphs from compiled binaries without the Ghidra GUI.

Use when executing implementation plans with independent tasks in the current session

Validate UI in the browser for layout, responsiveness, and interaction issues.

Use for new surfaces, redesigns, and ambiguous product requests: establish user tasks and business meaning. Built-screen audits and visual polish are separate skills.

Create story examples for components. Use when writing stories, creating examples, or demonstrating component usage.

Implementing custom elements using GPUI's low-level Element API (vs. high-level Render/RenderOnce APIs). Use when you need maximum control over layout, prepaint, and paint phases for complex, performance-critical custom UI components that cannot be achieved with Render/RenderOnce traits.

Generate CodMate hook drafts from requirements.

Reviews PRs or code changes using multiple AI models, synthesizes findings, and interactively applies fixes. Uses GPT 5.2, Gemini 3 Pro, and Opus 4.5 for diverse perspectives.

Run Semgrep static analysis scan on a codebase using parallel subagents. Supports two scan modes — "run all" (full ruleset coverage) and "important only" (high-confidence security vulnerabilities). Automatically detects and uses Semgrep Pro for cross-file taint analysis when available. Use when asked to scan code for vulnerabilities, run a security audit with Semgrep, find bugs, or perform static analysis. Spawns parallel workers for multi-language codebases.

Ensure all communication matches brand voice and tone guidelines. Use when creating marketing copy, customer communications, public-facing content, or when users mention brand voice, tone, or writing style.

Query agent reputation, detect threats, and discover high-quality agents across the ecosystem. Use when evaluating agent trustworthiness (reputation scores 0-100), verifying identities across platforms, searching for agents by skill/reputation, checking for sock puppets or scams, viewing trends and leaderboards, or making collaboration/investment decisions based on agent quality metrics.

Matches visualization types to data questions and creates narrated reports that highlight insights and recommend actions. Covers chart selection (comparison, trend, distribution, relationship, composition, geographic), perceptual best practices, and narrative reporting (headline, pattern, context, meaning, action). Use when analyzing data for patterns, building dashboards, presenting metrics, monitoring KPIs, or when user mentions "visualize this", "what chart should I use", "create a dashboard", "analyze this data", "show trends", "report on".

Session context loader. Reads the latest daily handoff to restore project context without redundant file exploration. Three-layer loading: CLAUDE.md (stable) → daily handoff (session state) → git verification (ground truth). Degrades to file exploration when no handoff exists. CQRS read-side pair with project-scribe.

Build entity triplets and brand associations that LLMs recognize and cite consistently.

Implement resilient Symfony async workflows with idempotency, retries, and operational visibility. Use for rate limiting tasks.

Automated PR review cycle for RPIV orchestration. Waits for Gemini reviews, assesses feedback, implements fixes, and manages the review loop. Used by Planner agent after RPIV opens PRs. This skill is triggered automatically by RPIV when entering the PR Review phase.

> This module is the `novel-harness`'s **specialized review submodule**. > Responsibility: analyze the chapter-level reading experience — whether climax intervals are reasonable, whether hook density is sufficient, whether satisfying-beat distribution is even, and whether chapter length is stable.