Coding & Dev Tools

open-mercato Skill

Open Mercato repo-local extension of the shared `om-prepare-issue` skill (installed from open-mercato/skills into .agents/skills/). Adds the --enterprise spec scope, duplicate-spec checks across both spec trees, and the om-implement-spec/om-auto-fix-issue pickup path.

open-mercato Skill

Open Mercato repo-local extension of the shared `om-auto-review-pr` skill (installed from open-mercato/skills into .agents/skills/). Makes the review GitHub-checks-first (local validation only as a narrow fallback) and keeps this repo's stricter verdict rule (Medium findings request changes).

open-mercato Skill

Open Mercato repo-local extension of the shared `om-followup-issue-from-pr` skill (installed from open-mercato/skills into .agents/skills/). Pins design-doc detection to .ai/specs/ (+ enterprise/), keeps the `enterprise` label, and hands specs off to om-implement-spec.

ferroxlabs Skill

IJFW efficiency layer -- smart output, routing, context discipline. Always active. Trigger: any IJFW interaction. Disable: 'ijfw off' or 'normal'.

ferroxlabs Skill

Terse conventional commits. Trigger: commit, git commit, /ijfw-commit

ferroxlabs Skill

One-line code review comments. Trigger: review, code review, PR review, /ijfw-review

ferroxlabs Skill

Check for and apply IJFW updates safely. Trigger: 'update ijfw', 'upgrade', 'latest version', 'is there a new version', /update

ferroxlabs Skill

Session handoff generation and loading. Trigger: session end, context full, /handoff

ferroxlabs Skill

Use when the user says: 'compute', 'crunch this', 'analyze logs', 'aggregate the data', 'run a script', 'dedupe', 'count by', 'top N', or any data-shaping ask. Replaces dumping raw data into context with sandboxed script execution.

ferroxlabs Skill

Compress memory/context files into terse form. Trigger: /compress, compress file

ferroxlabs Skill

Challenge decisions, surface counter-arguments, flag assumptions. Trigger: 'should I', 'is this right', 'critique', 'poke holes', 'second opinion', 'devil's advocate'. Auto-fired by ijfw-intent-router.

archibate Bundle

Use this skill when need weather forecast in Chinese locations

archibate Bundle

This skill should be used when using UV (Python package manager), needs to set up Python virtual environments, install/manage Python CLI tools, migrate from pip/pipx to UV, Python version management, or troubleshoot UV-related issues, or when about to write any `uv run`, `uv add`, or `uv sync` command.

archibate Skill

Use when interacting with TUI applications, or running processes that require stdin interaction. For non-interactive long-running tasks, prefer pueue instead.

archibate Skill

Use this skill to send notifications to the user. TRIGGER when user says "notify me", "send notification", "alert me".

archibate Skill

Skill to query and update a mistake notebook to memorize mistake. This skill should be used when: 1. Encountered a severe problem, needs to recall for historical problem solving experience 2. After fixed a severe mistake with a clear solution valuable to be recorded 3. User requested 'Do not make this mistake again'

pyfagorass Bundle

Transcribe audio files to text with optional diarization and known-speaker hints. Use when a user asks to transcribe speech from audio/video, extract text from recordings, or label speakers in interviews or meetings.

pyfagorass Skill

Downloads videos from YouTube and other platforms for offline viewing, editing, or archival. Handles various formats and quality options.

pyfagorass Skill

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation

pyfagorass Skill

Diagnoses Qdrant production issues using metrics and observability tools. Use when someone reports 'optimizer stuck', 'indexing too slow', 'memory too high', 'OOM crash', 'queries are slow', 'latency spike', or 'search was fast now it's slow'. Also use when performance degrades without obvious config changes.

aifinlab Skill

当用户需要进行客户适当性审核、投资者适当性匹配、产品准入审核、风险等级匹配、合规审查、销售前审核、材料完整性检查、投资经验与风险承受能力核验、适当性话术整理、审核意见撰写或形成标准化审核结论时使用。适用于金融机构、财富管理、券商、基金销售、私募、资管、保险及其他需要进行客户与产品适配审核的场景。

aifinlab Skill

IPO材料初审助手，适用于投行、券商、律所、会所及企业上市准备场景。 以下情况请主动触发此技能： - 用户提供了招股说明书、申报稿、问询函回复、法律意见书等文字，问"帮我看看"、"有没有问题" - 用户描述了材料中的某个疑问，问"这个表述有问题吗"、"这样写会被问到吗" - 用户需要整理问题清单、初审意见、自查报告 - 用户准备回复交易所问询函，需要梳理要点或材料框架 - 用户说"这里数据对不上"、"这两个地方说法不一样"、"这段披露够不够" 不要等用户明确说"IPO初审"——只要涉及上市申报材料的审核、检查、一致性核验、问询回复，就应主动启动此技能。

aifinlab Skill

当用户需要回答投顾场景中的客户问题、整理投顾问答口径、输出标准回应、解释市场波动、说明产品配置逻辑、解答申赎疑问、处理客户异议、形成服务话术或统一团队答复口径时使用。适用于财富管理、基金投顾、券商投顾、私行及客户服务场景。

aifinlab Skill

当用户需要快速解读公司财报、季报、年报、半年报、业绩快报、业绩预告或财务更新时使用。适用于帮助用户在短时间内提炼核心业绩变化、判断是否超预期、拆解增长和利润驱动、识别一次性因素、总结管理层口径、评估市场影响，并生成适合投研讨论、群内快评、领导汇报或研究纪要的财报点评。

aifinlab Skill

将多维经营分析结果整合为管理层可读的结构化摘要，支持经营分析会汇报材料的快速生成

aifinlab Skill

当用户需要分析客户持仓结构、诊断资产配置问题、识别风险暴露、评估集中度、解释收益表现、提出调仓建议、输出持仓点评或形成客户沟通材料时使用。适用于基金投顾、财富管理、券商投顾、私行及客户复盘场景。

aifinlab Skill

当用户需要准备尽调问题清单、访谈提纲、管理层问答、业务核查问题、财务核查问题、法务尽调问题、客户供应商访谈问题或项目尽调框架时使用。适用于投行、投资、并购、股权投资、风控及项目筛选场景。

aifinlab Skill

对银行核心经营指标（存款、贷款、收入、利润等）的变动进行多维归因拆解，定位增长或下降的关键驱动因素

aifinlab Skill

募资文件对比助手，适用于合规审查、内部复核、投资决策、版本管理等场景。 以下情况请主动触发此技能： - 用户提供了两份或多份文件/文本，需要找出差异（即使没有说"对比"） - 用户说"新版改了哪些"、"帮我看看变化"、"和旧版有什么不同"、"哪里有调整" - 用户提到：募集说明书/基金合同/合伙协议/定增方案/资管计划/推介材料的修订、更新、调整 - 用户描述了某些条款变化，需要判断影响（即使只有一份文件） - 用户需要整理成合规报告、审批材料、内部复核意见 - 用户问"这个变化重要吗"、"需要重新报备吗"、"这样改合规吗" 不要等用户明确说"文件对比"——只要涉及募资/基金/资管/融资文件的版本变化、条款调整、差异识别，就应主动启动此技能。

aifinlab Skill

对银行多个分行、支行或网点的经营指标进行横向对比与排名分析，识别领先和落后机构及其差异原因

entropy-r Bundle

基于已确认需求和仓库事实，为涉及跨模块、公共接口、数据流、持久化或迁移决策的软件变更生成复杂度受控、可验证且决策完整的工程方案。适用于“02 实现方案设计”“第二阶段”“把需求转成工程方案”和中大型项目设计；不用于可由局部修改完成的简单任务、正式 Ticket 拆分或代码评审。

entropy-r Bundle

Use when Codex needs to turn an incomplete or unstructured product/software request into a structured requirement intake, actively ask clarifying questions, identify missing or ambiguous information, and prepare a stable input package for a later solution-design stage. Triggers include Chinese requests like “01 需求理解与澄清”, “需求理解与澄清”, “帮我澄清需求”, “第一阶段”, “需求还不完整”, “先问我问题”, or requests involving text, image-dialog descriptions, document excerpts, or vague feature ideas.

entropy-r Bundle

Adapt a customer document template and generate an outline using the source-analysis result from a technical agreement. Use when Codex receives a customer template plus the output of doc-source-analysis and needs to produce a document directory, chapter outline, section writing guidance, missing section suggestions, or template-fit plan for software design documents, requirement specifications, test reports, or user manuals.

entropy-r Bundle

按 SKILL.md 明确列出的规则识别并改写写作中的 AI 痕迹。只能处理清单内的问题；未命中规则的文字必须原样保留，也不能改变文章框架。适用于写作完成后的成稿清理。 Remove AI writing tells using an explicit whitelist of rules; leaves unmatched text untouched.

entropy-r Bundle

Proofread software project document content against user-provided references. Use when Codex needs to check requirement specifications, design documents, test reports, user manuals, or similar documents for factual consistency, terminology consistency, logical completeness, boundary conditions, requirement traceability, acceptance risk, source conflicts, and precise replacement suggestions after document formatting has been handled separately.

entropy-r Bundle

Guardrails for Codex when running commands or editing files on Windows PowerShell. Use when working in Windows shells, especially with Chinese paths or filenames, Chinese logs or script output, text encoding, JSON passed through PowerShell, quoting and escaping, path safety, deletion or move operations, cross-shell command composition, PowerShell startup failures, Python or Node scripts affected by Windows locale encodings, and repository maintenance where PowerShell behavior can corrupt text or affect unintended paths.