Top Agent Skills

Use when implementing loki functionality with production-grade patterns and safeguards.

Use when implementing perl functionality with production-grade patterns and safeguards.

Use when implementing trpc functionality with production-grade patterns and safeguards.

Expert email writing skill for professionals and individuals. Use this skill whenever the user wants to write, draft, improve, or get help with ANY email — including cold outreach, follow-ups, complaints, negotiations, thank-you notes, difficult conversations, reconnection emails, or any other professional or personal email. Also trigger for email marketing requests: promotional offers, newsletters, welcome emails, re-engagement campaigns, or clickbait-style emails. Trigger for: "help me write an email", "draft an email", "how should I phrase this email", "I need to reach out to someone", "I need to follow up", "I want to send a thank you", "I have a complaint to write", "write a newsletter", "create a welcome email", "write a promotional email", "re-engage my subscribers", or any situation where someone needs to communicate via email. Do NOT limit this to obvious requests — if the user describes a situation where an email would be the right output, use this skill proactively.

Expert content repurposing skill for creators and professionals who want to get more out of content they've already created. Use this skill whenever the user wants to turn one piece of content into multiple formats — for example: turning a blog post into social media posts, converting a YouTube video or podcast into a newsletter or article, transforming a long-form piece into short clips or quotes, or adapting any content for a different platform or audience. Triggers: "repurpose this", "turn this into a post", "make a LinkedIn post from this", "convert this blog to a tweet thread", "I have a video/podcast, can you make it into a newsletter", "get more out of this content", "adapt this for [platform]", or any request to transform existing content into a new format. Do NOT limit to obvious repurposing requests — if the user shares content and mentions another platform or format, use this skill.

Foundational C programming patterns covering type system, memory management, pointers, arrays, preprocessor, compilation, concurrency (pthreads, mutexes, atomics), serialization (binary, JSON, struct packing), and testing (Unity, CMocka, Check). Complete 8/8 pillar coverage. Use when writing C code, understanding manual memory management, working with system-level programming, or needing guidance on which specialized C skill to use. This is the entry point for C development.

Provides foundational patterns for building type-safe frontend applications with Elm, covering The Elm Architecture, core language features, and Elm's approach to functional web development.

Foundational Java patterns covering core syntax, object-oriented programming, generics, collections, streams, lambdas, and modern Java features. Use when writing Java code, understanding the type system, working with collections/streams, or needing guidance on which specialized Java skill to use. This is the entry point for Java development.

GitHub App development guide for building custom integrations. Use this skill when creating a GitHub App, building webhook handlers, implementing GitHub API integrations, developing PR/Issue automation apps, or deploying GitHub Apps to Cloudflare Workers.

Master authentication and authorization patterns including JWT, OAuth2, session management, and RBAC to build secure, scalable access control systems. Use when implementing auth systems, securing APIs, or debugging security issues.

Foundational .NET patterns covering runtime, project structure, dependency injection, configuration, metaprogramming, and cross-platform development. Use when working with .NET projects or CLI tools. This is the entry point for .NET development.

Foundational Python patterns covering core syntax, idioms, type hints, testing, and modern tooling. Use when writing Python code, understanding Pythonic patterns, working with type hints, or needing guidance on which specialized Python skill to use. This is the entry point for Python development.

Convert Java code to idiomatic C++. Use when migrating Java projects to C++, translating Java patterns to idiomatic C++, or refactoring Java codebases. Extends meta-convert-dev with Java-to-C++ specific patterns.

Develop reusable library code across languages. Use when designing public APIs, organizing modules, managing versioning, or creating utility libraries. Provides foundational patterns that language-specific *-lib-* skills extend.

Bidirectional conversion between Roc and Scala. Use when migrating projects between these languages in either direction. Extends meta-convert-dev with Roc↔Scala specific patterns. Use when migrating Roc projects to JVM/Scala, translating pure functional patterns to object-functional hybrid, or refactoring Roc codebases. Extends meta-convert-dev with Roc-to-Scala specific patterns.

Convert Objective-C code to idiomatic Swift. Use when migrating Objective-C projects to Swift, translating Objective-C patterns to modern Swift, modernizing legacy codebases, or establishing gradual migration strategies with interop. Extends meta-convert-dev with Objective-C-to-Swift specific patterns including bridging and @objc attributes.

Bidirectional conversion between Elm and Haskell. Use when migrating projects between these languages in either direction. Extends meta-convert-dev with Elm↔Haskell specific patterns. Use when migrating Elm frontend code to Haskell, translating Elm patterns to idiomatic Haskell, or refactoring Elm codebases. Extends meta-convert-dev with Elm-to-Haskell specific patterns.

Develop OpenFeature SDK implementations from the specification. Use when implementing the OpenFeature spec in a new language, extending existing SDKs with new features, building custom providers, or contributing to official SDK repositories.

C++-specific library development patterns. Use when creating C++ libraries, designing header-only vs compiled libraries, configuring CMake for library targets, managing ABI stability and versioning, integrating with Conan/vcpkg, documenting with Doxygen, or testing with GoogleTest/Catch2. Extends meta-library-dev with C++ tooling and idioms.

Bidirectional conversion between Erlang and Fsharp. Use when migrating projects between these languages in either direction. Extends meta-convert-dev with Erlang↔Fsharp specific patterns.

Look up and apply the cooked, indexed proof playbook built from proof-material. Use when the user is proving a theorem, planning a proof, seeking reusable proof moves by task shape, assumptions, tags, desired move, or material source ID, or adapting cooked local moves and macro strategies with public source attribution.

Review Markdown prose against agent-style's 21 rules. `/style-review FILE` audits the file and (on user confirm) writes a polished copy at `FILE.reviewed.md`. `/style-review A.md B.md` A/B-compares two drafts and emits a per-rule delta table. Complements the generation-time rules installed via `agent-style enable <tool>`.

Store and inspect source-indexed proof material extracted from papers, notes, PDFs, or LaTeX sources. Use when Codex needs stable material IDs, source locations, reusable abstractions, proof skeletons, tags, and public attribution before any proof ideas are cooked into proof-usage.

21-rule prose review workflow for technical and academic Markdown/LaTeX excerpts. Use when the user asks for muxing-style-review, wants a complete agent-style check combining check-human.md with agent-style audit-only evidence, wants a compact-rule check, wants a DeepSeek/Codex manuscript draft checked for clarity, evidence, citation discipline, terminology, AI-tell prose, or wants an A/B draft comparison. Prefer a dedicated review subagent when available for non-trivial reviews. Use the sibling style-review skill for the upstream post-hoc audit/polish workflow.

Chrome-dependent academic paper fetching and BibTeX curation. Use when the user provides a planNN.md, paper list, citation list, or literature plan and wants Codex to use Google, Google Scholar, publisher pages, DBLP, arXiv, and the user's Chrome-accessible permissions to fetch the best available PDFs and create or update references.bib.

Production Role Session

Project Status Overview

Execute Stored Procedures on a Fabric Warehouse

Execute SQL on a Databricks SQL Warehouse

Inventory Items in a Fabric Workspace

Execute a Fabric Data Pipeline

Read/Write Permission Test on a Fabric Warehouse

Delete a Fabric Workspace (with safety hook)

Turns a rough request into working software via a governed build loop: align first, plan, then iterate one task at a time with deterministic backpressure and holdout-scenario judging.

AI and machine learning workflow covering LLM application development, RAG implementation, agent architecture, ML pipelines, and AI-powered features.

Rewrite the previous response more briefly while preserving the substance.

Interact with GitHub issues - create, list, and view issues.

Fast in-memory DataFrame library for datasets that fit in RAM. Use when pandas is too slow but data still fits in memory. Lazy evaluation, parallel execution, Apache Arrow backend. Best for 1-100GB datasets, ETL pipelines, faster pandas replacement. For larger-than-RAM data use dask or vaex.

Scanpy is a scalable Python toolkit for analyzing single-cell RNA-seq data, built on AnnData. Apply this skill for complete single-cell workflows including quality control, normalization, dimensionality reduction, clustering, marker gene identification, visualization, and trajectory analysis.

Compare AI agent-made UI variants locally in a real app, then keep one and clean up unused temporary code.

Human-in-the-loop web studio to tune AI-generated output by eye. Stand up a local interactive studio (sliders, pickers, drag handles) or an inline edit/highlight/comment annotation studio for prose & media, instead of guessing values or shipping a static comparison grid.

Audit a component or page for accessibility issues and fix them

Design user flows and navigation structure following proven UX patterns

Ask which skill or flow fits your situation. A router over the user-invoked skills in this repo.

Use when the user says "task:", "fix:", "todo:", "log task:", "log fix:", "run task:", "run fix:", or asks to log, run, list, or manage project tasks. Also use when asked to generate or update a changelog from completed tasks.

Use when work requires inter-agent communication, cross-layer coordination, or configurable autonomy with Claude Code Agent Teams

Use to build a data-subject erasure plan for support conversations — GDPR right to erasure, CCPA deletion, right to be forgotten. Trigger for "delete this customer's data", DSR or SAR erasure request, GDPR deletion from our helpdesk, redacting personal data from tickets, or working out what can and cannot be erased.

Use to compute and monitor regulated complaint deadlines — acknowledgement, holding response, final response — against business-day calendars, and to find cases about to breach. Trigger for "are we meeting our complaint deadlines", "complaint SLA breaches", "final response deadline", "which complaints are about to breach", complaint ageing reports, or building a complaints deadline monitor.

Use to test whether a support control actually operates, the way an auditor would test it, rather than confirming it exists on paper. Trigger for "test our controls", "does this control actually work", second-line assurance over support, control effectiveness testing, preparing for an internal audit, or a control that passed review and then failed in practice.

Use to find the activation blockers visible in support contacts from new customers, before they show up in retention numbers months later. Trigger for "why aren't new customers activating", "what do new customers contact us about", onboarding drop-off, time-to-value analysis, first-90-days support experience, or a retention problem that traces back to activation.

Use to build a defensible cost per contact and cost to serve by driver, channel or customer segment, for automation business cases and budget decisions. Trigger for "what does a ticket cost us", "cost per contact", "what would automating this save", "cost to serve by segment", building a business case for a bot or a process change, or a savings claim that looks too good.

Use to quantify what support quality failures cost — rework, credits, escalation handling, complaints and churn — with the causal limits stated. Trigger for "what do our quality failures cost", "cost of rework in support", "business case for improving quality", "how much is bad support costing us", or a quality investment that needs a number attached.

Use to run customer support communications during an outage or major incident — banners, macros, status page, volume planning, and postmortem inputs. Trigger for "incident comms playbook", "support during outage", status page and help centre banner, macro for widespread issue, staffing surge, or freeze non-incident work.

Use to analyse touch time vs elapsed time across the resolution path — queue wait, customer wait, back-office wait, and where elapsed time hides real delay. Trigger for "flow efficiency", "touch time vs elapsed time", "where is the time going", "customers waiting on us", back-office bottlenecks, idle time in cases, or end-to-end time much longer than handle time.

Use to find duplicate or repeated support conversations and produce a reviewable merge plan. Trigger for "find duplicate tickets", "customers contacting us twice about the same thing", deduplicating a helpdesk, cleaning up ticket volume, or measuring how much of your contact volume is the same issue submitted more than once.

Use to audit a CX dashboard for tiles nobody acts on and metrics nobody agrees on — assign an owner per tile, apply kill criteria, and run an actionability test before adding or keeping a chart. Trigger for "clean up the dashboard", "too many metrics", "which KPIs should we keep", dashboard audit, metrics review workshop, "nobody looks at this anymore", or before a BI migration or exec-pack refresh.

Use to verify that the chain from conversation to evaluation to decision is complete and reconstructible, so a QA score or a customer outcome can be explained months later. Trigger for "can we prove why this decision was made", "is our audit trail complete", "reconstruct how this score was produced", evidencing an AI-assisted decision, or an auditor asking how a conclusion was reached.

Use to design a safe pipeline that writes derived attributes back into a helpdesk — risk scores, segments, health flags — without corrupting agent workflows or creating an unexplainable feedback loop. Trigger for "push scores back into Zendesk", "sync warehouse attributes to the helpdesk", "write our churn risk onto the ticket", reverse ETL for support, or an automation that started firing on a synced field.

Use to measure whether a support bot, AI agent, or self-service channel actually reduces contact volume, and to audit a vendor's containment or deflection number. Trigger for "what's our real deflection rate", "is the bot working", "our containment rate is 70% but tickets haven't dropped", automation ROI, self-service savings, AI agent resolution rate, or designing a holdout test for a support bot.

Use to write CX commentary that makes a number decision-ready — what moved, why (decomposed), so what, do what — separating noise from signal and stating confidence honestly. Trigger for "write the commentary", "explain this trend", "what should we say about the drop", weekly or monthly narrative, board pack prose, "why did QA fall", metric movement write-up, or when a report has charts but no story.