Skill Claude CodeCodex
Bootstrap GitHub Copilot CLI for GBB workflows: 6 recommended MCP servers (mslearn, Azure, Playwright, context7, tavily, mem0), settings.json baseline (model, sessionSync, allowedUrls, trustedFolders), work-iq plugin family for Microsoft staff, autoApprove guardrails, and a fresh-machine bootstrap procedure the agent…
Skill Claude CodeCodex
Deploy Foundry hosted agents using GitHub Copilot SDK (GHCP) with BYOK authentication and Invocations protocol. Alternative to MAF (Agent + FoundryChatClient + ResponsesHostServer). Use when you need streaming SSE output, progressive skill discovery, or long tool loops (>120s) that hit Foundry gateway timeout with…
Skill Claude CodeCodex
Discover AI Apps GBB IP catalog via MCP server. Search, list, filter, and inspect IPs (intellectual property assets) including metadata, thumbnails, and READMEs. USE FOR: find IP, search demos, list IPs, catalog, what demos are available, find solutions by service/pattern, get IP details, browse catalog, discover…
Skill Claude CodeCodex needs its repo
Headless Azure OpenAI PAYGO-vs-PTU cost analysis and sizing report. Wraps the analysis library of aiappsgbb/ptu-paygo-mix (NOT its Streamlit UI). Supports CSV token-usage input and Log Analytics queries via vendored KQL against AzureMetrics. Produces a markdown report, JSON data, and charts. Bundles a synthetic-data…
Skill Claude CodeCodex needs its repo
Build job-backed MCP servers for Foundry when the control plane must stay responsive and execution belongs in ACA Jobs. USE FOR: MCP Tasks, SEP-2663, ACA Jobs, durable result claims, callbacks, external job orchestration, and shared-image worker handoff. DO NOT USE FOR: producer-side MCP hosting (use foundry-mcp-aca)…
Skill Claude CodeCodex
Use when adopting native Azure AgentOps for one existing agent. USE FOR: AgentOps, agent operations, per-agent operational readiness, Doctor, results.json/evidence.json review, release evidence, baseline decisions, trace-to-regression, workflow analysis, and standalone or Threadlight prerequisites. DO NOT USE FOR…
Skill Claude CodeCodex needs its repo
Build Microsoft Agent Framework Python agents with createharnessagent: plan/execute modes, persistent todos, context compaction, session/file memory, approval UX, recovery, and opt-in file access, skills, background agents, shell, and bounded looping. Covers actual defaults, internal provider/middleware ordering…
Skill Claude CodeCodex
Design and improve task-centered web experiences: executive HTML reports, analytical comparisons, catalogs, dashboards, interactive tools, documentation, informational sites and navigable HTML presentations. Turn complex content into clear views, useful visual explanations and working navigation before styling. USE…
Skill Claude CodeCodex
Execution memory and bounded recovery without planning bureaucracy. USE FOR: long-running or multi-phase work, repeated failed attempts, stalled tools, plan changes, resumption after compaction, child-session coordination, handoffs, stopping rabbit holes, reviewing real progress, or maintaining an execution ledger. DO…
Skill Claude CodeCodex
Read-only maintenance doctor for an existing GitHub Copilot App or CLI setup. USE FOR: /copilot-doctor, setup health, broken MCP launchers, runtime and startup diagnosis, stale paths or commands, plugin and skill duplicates, trigger overlap, instruction conflicts, configuration drift, comparing local health scans, and…
Skill Claude CodeCodex
Unreleased draft for service-native voice-based prompt agents in Microsoft Foundry Agent Service public preview. Uses a standalone Azure AI Projects voice client, not a hosted speech pipeline. USE FOR: Foundry voice agent, kind voice, prompt-voice, managed voice model, voice agent greeting, synthetic audio turn, voice…
Skill Claude CodeCodex
Create and refine polished, evidence-backed HTML presentations using continuous visual narratives and purposeful exploration. Use for HTML or browser-based briefings, pitches, retrospectives and scrollytelling presentations. Deliver distinctive responsive web design suited to guided talks and independent reading. USE…
Skill Claude CodeCodex
Reduce cost per successful task, TPM pressure and context size for Responses API apps and agents. Ordered ladder: prompt/tool-output diet, deferred tool loading with toolsearch + deferloading (gpt-5.4+), cache-friendly prefixes, allowedtools, reasoning effort/verbosity/maxoutputtokens, compaction, model right-sizing…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: