Latest Agent Skills

Create, validate, and consume Google's Open Knowledge Format (OKF) bundles — YAML-frontmatter Markdown files with type / title / description / resource / tags / timestamp fields for portable, interoperable AI-agent knowledge sharing. OKF formalizes the LLM-Wiki pattern into a vendor-neutral open specification so any producer can write and any agent can consume without translation. Routes: use `llm-wiki` for raw source capture + vault maintenance, `obsidian` for Obsidian-vault workflows, `graphify` for durable committed graphs, `scrapling` for web-content extraction into OKF docs. Triggers on: okf, open knowledge format, knowledge bundle, okf document, llm wiki standard, knowledge atom, agent context format, okf frontmatter, okf bundle, knowledge interoperability.

Drive mex (`mex-agent`), persistent project memory and code graphs for AI coding agents. One command scaffolds a living wiki, builds a deterministic code graph, and installs a project anchor file (CLAUDE.md, root AGENTS.md, .cursorrules, .windsurfrules, copilot-instructions.md, or .opencode/opencode.json) that your agent auto-loads as a standing rule document. Use when the user wants to `mex setup` a new project, build a symbol-grounded wiki, keep knowledge connected to implementation, route relevant context to agents, or run drift detection (`mex check`, `mex sync`). Triggers on: "mex setup", "project memory", "code graphs", "codebase documentation", "drift detection", "agent memory", "structured scaffolds", "architectural context", "living wiki", "project anchor file".

AI-era curl for agent web fetching, structure discovery, and deterministic structured extraction with token-budgeted output.

分析故事人物关系，识别类型、特点及发展变化。适用于深度理解人物关系网络、分析关系对情节推动作用、为情节设计提供关系支撑

基于大情节点，深入分析并生成详细情节点描述及情节发展说明。适用于细化故事大纲、指导剧本写作、分析情节内在逻辑

基于故事文本分析人物特征，生成详细的人物小传。适用于深度理解故事人物设定、为演员塑造角色提供参考、为剧本创作建立人物档案

分析故事情节点，识别关键情节与转折点。适用于深度分析情节结构、评估情节发展有效性

整合多情点分析结果为综合报告，通过去重、分类、排序、总结生成高质量分析。适用于整合多分析源、生成统一报告

分析故事题材类型，提炼创意元素与故事特色。适用于识别题材定位、提炼创意、分析市场受众

全面分析故事五元素：题材类型、故事梗概、人物小传、人物关系、大情节点。适用于深度分析故事、剧本改编准备、故事开发文档制作

整合智能体输出结果为结构化最终输出，支持多种格式。适用于整合分析报告、生成结构化报告

专注于竖屏短剧知识库查询和信息检索，提供专业知识服务。适用于查询剧本桥段、高能情节、创作技巧、商业运营等专业知识

专业评估故事大纲，从市场潜力、创新属性、内容亮点维度判断评分。适用于故事大纲质量评估、IP改编潜力判断、项目立项决策

阅读理解小说文本，总结成流畅的故事大纲。适用于小说初筛选、生成500-800字故事大纲

基于故事文本提炼主要情节与要点，生成完整故事梗概。适用于快速了解故事内容、剧本改编、项目推介

分析已播剧集，包括信息获取、拉片分析、五元素分析及联网搜索。适用于深度分析剧集创作技法、学习情节设计

智能截断文本，保持内容完整性。适用于小说文本预处理、确保文本不超过指定长度限制

严格细致判断与评分故事文本，从市场潜力、创新属性、内容亮点维度分析质量。适用于小说初筛选、多维度评估打分

Score Analyzer

阅读理解故事文本，总结人物、关系、情节，整理成流畅大纲。适用于快速了解故事核心、为剧本创作提供大纲基础

梳理故事主线，提炼并按发展阶段排列主要情节点。适用于快速掌握故事结构、制作大纲及剧本改编结构梳理

分析故事文本，提炼主要情节点并分析戏剧功能。适用于分析小说、剧本大纲、故事梗概等文本，识别关键转折点和情感节点

使用联网搜索获取最新市场信息和行业动态。适用于获取竖屏短剧市场趋势、热门题材分析、成功案例研究

从思想性、艺术性、观赏性三维度评估影视剧本并打分。适用于剧本开发质量评估、修改方向确定、项目立项前审查

依据竖屏短剧评估标准，从核心爽点、故事类型等维度评估打分。适用于评估故事改编为竖屏短剧的潜力、分析市场竞争力

调用generateTreeMind工具创建可视化思维导图，返回图片URL和编辑链接。适用于故事结构可视化、情节关系图谱展示

协调情节点戏剧功能分析流程，管理文本预处理、并行分析、结果整合。适用于长文本的情节点和戏剧功能分析、需要结构化分析报告的场景

梳理IP网络信息并进行多维度评估打分。适用于评估小说、剧本等IP的改编价值、分析市场潜力和创新属性

编排和协调大情节点与详细情节点的一键生成工作流。适用于长篇故事的结构化分析、需要模块化智能体协作的复杂任务

制定竖屏短剧专业策划方案，含情绪价值分析、黄金三秒钩子设计、三幕式结构规划等。适用于短剧项目前期策划、商业化方案设计、创作指导

创作竖屏短剧剧本，包括宏观建构、剧本创作、精准优化、创意发想。适用于从零开始创作短剧、优化现有剧本、设计故事大纲和悬念钩子

智能截断文本，保持内容的完整性和语义连贯性。适用于长文本预处理、确保文本不超过指定长度限制

处理和分析用户上传的文件引用，提取关键信息并提供应用建议。适用于解析@符号引用、自然语言引用、提取短剧策划相关信息

将文本分割成指定大小的块，保持语义完整性。适用于长文本批处理、将文本分割为可处理的片段

Inspect screenshots, photos, diagrams, image paths, and image URLs when the task depends on visible content. Use when the prompt lacks actual image content, native inspection fails, or the user requests inspect_image; prefer the MCP tool, then the installed CLI.

Query and display the current API provider balance and recent usage.

Summarize a contributor's Git history, a provided work log, or both into a concise, review-friendly self-evaluation for quarterly, semi-annual, or promotion cycles.

Summarize each day's Git activity into a concise daily work log, for a single date or a range. Uses the current repository, optional configured work projects, or paths named in conversation; configuration is never required.

Refactor changed code to reduce duplication, complexity, and wasted work. Use for diffs, commit ranges, PRs, paths, staged changes, or working-tree changes.

Review code changes for bugs, regressions, convention violations, and high-value cleanup opportunities. Use for diffs, commit ranges, hosted PR/MR URLs, branches, paths, staged changes, or working-tree changes.

Generate a Conventional Commits message from staged changes and wait for confirmation before committing. Use when the user asks to commit or generate a commit message.

Work ZenTao bugs/tasks end to end, manage task lifecycle and hours, or read a linked story as development context: fetch details, handle the user's request, verify changes, and ask before committing or writing back to ZenTao. Supports single items and sequential batches. Use when the user references ZenTao (禅道) bugs, tasks, stories, requirements, task status, or task hours.

Use when designing or auditing the evaluation of an ASE (IEEE/ACM Automated Software Engineering) paper, covering real subject systems, fair runnable tool baselines, task-matched effectiveness metrics, ablations that isolate a learned component, oracle and correctness validation, contamination-aware LLM handling, and provenance for mining.

Use when designing or auditing experiments for an ACL paper, covering tuned LLM baselines, multi-dataset and multilingual evaluation, statistical significance and variance, human evaluation with agreement reporting, contamination and prompt-sensitivity controls, ablations, and error-analysis expectations in NLP reviewing.

Use when planning the calendar for a WSDM submission - the single annual cycle from August abstract and paper deadlines through October notification to the February conference, backward milestone planning, the September short-paper fallback, alternate entrances (demos, WSDM Cup, doctoral consortium), and coordination with sibling-venue deadlines.

Use when planning a WACV campaign calendar across the two-round model, from track and round choice through the Round 1 submission, rebuttal, and Accept/Revise-and-Resubmit/Reject decision, into the Round 2 revision and decision, camera-ready, and the winter conference — with the branch points that a single-deadline plan misses.

Use when planning a VLDB project on the PVLDB rolling calendar, covering month-picking strategy, the volume-to-conference mapping and acceptance cutoff, budgeting a possible three-month revision, coordinating author caps across a group, and backward planning from the 25th-abstract / 1st-paper rhythm.

Use when auditing an IEEE VIS full-paper submission for PCS readiness, covering the abstract-then-paper two-deadline structure under the VGTC society, the IEEE VGTC/TVCG 9+2 page budget, author-optional double-blind anonymization, the supplemental-material one-week window, and desk-reject triage before the AoE cutoff for a paper that will publish in IEEE TVCG.

Use when planning a UIST cycle end to end — the late-March abstract and paper deadlines, the ten-week build-freeze before video production, the rebuttal window, conditional acceptance, the July camera-ready, adjunct-track parallel lanes, and the November conference — with owners and risk buffers.

Use when auditing a UAI submission for OpenReview readiness, covering the February paper deadline, the 8-page main part with unlimited appendices in one PDF, the 15 MB file cap, the optional 50 MB supplementary ZIP, double-blind rules, the reviewer-volunteer agreement, dual-submission checks, and desk-reject triage before upload.

Use when planning a STOC (ACM Symposium on Theory of Computing) project calendar — backward planning to the early-November deadline, the STOC/FOCS alternating two-deadline year that theory groups schedule around, full-version and camera-ready milestones, and the June TheoryFest endpoint.

Use when planning a SOSP campaign across the annual cycle — back-scheduling from the spring abstract and paper deadlines, running the summer response and July notification, stacking the August camera-ready with artifact evaluation, and retargeting rejections across the SOSP/OSDI/EuroSys circuit.

Use when planning a SODA (ACM-SIAM Symposium on Discrete Algorithms) campaign end to end — backward-planning to the July AoE deadline, SODA's seat in the theory calendar between STOC and FOCS deadlines, the September rebuttal and October decision, satellite deadlines (ALENEX, SOSA), and the January conference.

Use when planning an ACM SoCC project timeline from venue fit through the two-round abstract+paper deadlines, the rebuttal, notification, camera-ready, and presentation, with backward-planning offsets for a cloud-systems measurement paper and honest handling of the two-rounds-per-year cadence and the no-cross-round-resubmission rule.

RSS Submission

Use when planning a POPL campaign calendar — backward-scheduling theory and mechanization from the July deadline, riding the October notification into the conditional-acceptance revision and artifact evaluation, landing the January conference, and retargeting across the SIGPLAN/PACMPL deadline wheel if July slips.

Use when planning an ACM PODS project timeline across its multiple submission cycles per year, from venue fit through the EasyChair abstract-then-paper deadlines, the 48-hour rebuttal, the accept/reject/revision decision and shepherded revision round, PACMMOD-track publication, the arXiv full version, and presentation — with backward-planning offsets for a theory paper and honest handling of cycle choice.

Use when planning an ACM PODC campaign end to end — running the single annual cycle backward from the February deadline through abstract registration, lightweight double-blind review, the late-April notification, the May camera-ready, and the arXiv full version — and coordinating the regular-paper-vs-Brief-Announcement decision along the way.

Use when deciding which pnas-* sub-skill to invoke next, or when sequencing a manuscript from significance test through reviewer rebuttal for PNAS (Proceedings of the National Academy of Sciences). Routes — it does not replace — the specialized skills.

Use after PNAS reviews arrive to triage the decision, prioritize experiments, and draft a point-by-point response that is respectful, evidence-led, and honest about limits. Do not run before the main text is actually revised.