uygnoey
- 294 skills
- 0 followers
- 4 repo stars
- 6 days ago last updated
- ▌ Inbound Buying Agent Design · uygnoey bundleDesign, build, and iterate an inbound "buying agent" that takes a prospect from "I want this for my company" to a purchase, a hand-off to a sales rep, or a quick answer - modeled on how Anthropic's sales team rebuilt inbound with Claude Managed Agents (beta) in September 2026. Covers where to deploy the agent (contact-sales and pricing pages, in-product, email), the three conversation outcomes, keeping the experience opt-in, a goal-first and short system prompt, a small tool set plus knowledge base, versioned prompt iteration with a staging agent and subject-matter experts in the loop, recommending the right (not the biggest) plan, and treating every agent-to-rep escalation as feedback. Use when a sales or go-to-market team cannot keep up with inbound demand, wants an always-on self-serve buying path, or is writing and tuning the prompt for such an agent.
- ▌ Idc Quanta Research Workflows · uygnoey bundleAnswer market and technology questions in Claude through the IDC Quanta MCP connector, following the October 2026 IDC announcement - competitive briefs, vendor shortlists delivered as Excel scorecards, comparing your own budget to IDC forecasts, and acquisition-impact analysis, each cited to the IDC document or dataset it came from. Also covers how IDC rolled Claude out internally (Claude Code for prototypes, parallel tests, and reviews; Cowork for plan drafts; CRM-export deal triage; a marketing content assistant). Use when a customer with IDC Quanta wants finished analysis instead of reading long reports, or when a research firm plans a company-wide Claude rollout.
- ▌ Domain Agent Builder · uygnoey bundleDesign a natural-language agent builder on top of the Claude Agent SDK, following the approach Cresta used for Conductor, its CX agent builder — the SDK as the general-purpose execution harness, your domain workflows, tools, and context layered on top, a blueprint → implementation → evaluation → optimization flow, explicit decisions about which workflows must be deterministic, a fixed build-task evaluation suite rerun on every model or framework change, and reusable memory artifacts shared as skills. Use when a team with deep domain expertise wants to turn it into a product that builds production-ready agents.
- ▌ Orchestrated Code Migrations · uygnoey bundleRun large code migrations (hundreds to thousands of pull requests across repos) with a tiered model setup in which a frontier orchestrator model does the upfront design — data models, API specs, algorithms — and directs up to dozens of smaller, cheaper worker models that make the file edits and run tests, while a human engineer steers and humans keep merges, deploys, and feature flags. Also covers picking a model and effort level per task from a benchmarked model × effort grid, recurring codebase cleanup passes, and safeguards that run outside the model loop. Based on how Block uses Claude Fable 5 with Opus and Sonnet workers. Use when planning or operating a multi-PR migration, choosing models for agent work, or setting guardrails for agent-written code.
- ▌ Design Coachable Agent Teammates · uygnoey bundleSet up AI agents as teammates on a shared work platform the way Asana does — structure unstructured work with Claude before agents act, give each agent a defined role, profile, scoped access bounded by the triggering user's permissions, and separate admins/editors who can commit to its shared memory from users whose feedback only applies to the current task, and keep agent work visible on shared tasks so reviewers can coach it. Use when designing human-agent teams, writing an agent's role profile, deciding who trains an agent, or turning a recurring team job into an agent workflow.
- ▌ Smb Run And Grow Workflows · uygnoey bundleRun a small business owner's week with Claude using the Claude for Small Business workflow pattern - onboarding (/smb-onboard), a Monday brief of cash, sales, pipeline, and overdue invoices (/monday-brief), after-hours lead response (/speed-to-lead), proposals from voice memos priced from past jobs (/proposal-builder), marketing content and review replies (/social-content-engine, /marketing-monday), month-end close with a packet for the accountant (/close-month), and gap fillers for missing connectors or steps (/build-connector, /build-agent). Every workflow starts in approval mode, some stop before the final step (payroll staged in Gusto), existing software permissions hold, and each workflow has a no-connector fallback from uploaded files. Use when an owner wants a weekly brief, lead replies, a proposal, a content calendar, or a month-end close.
- ▌ Agent Scale CI Test Selection · uygnoey bundleDesign or rescue a CI test impact analysis (test selection) service for agent-driven development, based on Anthropic's experience after CI jobs grew ~25x in six months. Covers the listener/selector architecture, why listener lag means stale selection, recognizing when quick fixes (bigger machine, per-package sharding, restarts) are buying less time each round, the stateless redesign (listener workers append to a journal in an in-memory store, a small consumer rolls it up into per-test history every few seconds, the selector reads that history), and the planning rules - assume 25x load within two quarters, design v0 for 10-20x perceived scale if budget allows, keep state out of the process, avoid single-instance critical services, and instrument services so Claude can diagnose and tune them (jobs in must equal jobs out). Use when CI is straining under agent-generated PRs, when a test selection service lags or hits memory limits, or when designing CI infrastructure for exponential growth.
- ▌ Frontier Model Eval Governance · uygnoey bundleEvaluate a new frontier model on real, verifiable domain tasks inside your own agent harness, validate surprising results, route models by evidence, and govern agent deployments with controls around the model (capability never grants authority). Follows the pattern Balyasny Asset Management (BAM) described in September 2026 for Claude Fable 5 and its BAMAgent platform. Use when deciding whether to switch a production model, designing an internal eval suite, setting model routing guidance, or defining the guardrails and human review for long-running agents in a regulated or high-stakes domain.
- ▌ Commitment Funded Partner Procurement · uygnoey bundlePlan and request the purchase of partner software (for example Snowflake capacity, Snowflake's inference API, or a committed Vercel plan) through Claude Marketplace using part of an existing Anthropic commitment, based on how CodeRabbit, Power Digital, and ThoughtSpot did it (September 2026). Covers spotting candidate tools the team already relies on alongside Claude, checking eligibility (limited preview - talk to the account team or submit a request), framing the purchase as already-approved budget rather than a new budget request, and the seller-side view for partners who want to sell through the marketplace. Use when a team with an Anthropic commitment wants to expand a data platform or infrastructure vendor without a new budget cycle, or when a vendor building on Claude asks how marketplace selling helps.
- ▌ Advisor Workflow Skills · uygnoey bundleRun a financial advisor's daily workflows with Claude the way the Claude for Financial Advisors plugin frames them - pre-meeting prep, post-meeting notes and follow-up, prospect intake, advisor onboarding, portfolio rebalance review, alternative investments brief, estate and tax brief, and compliance and AI policy - drawing on connected custodians, portfolio platforms, CRMs, planning and estate tools, and meeting capture. Claude gathers, summarizes, and drafts; it stages CRM updates and client communications for advisor approval, and leaves investment recommendations, client communications, and compliance determinations to human review. Use when an advisor or RIA wants a meeting brief, a drift explanation, a transcript-to-follow-up pass, an estate/titling check, an alts summary, a prospect handoff, or a marketing-language screen.
- ▌ Goal Driven Multi Thread Projects · uygnoey bundleSet up and steer a redesigned Claude Code project (beta, September 2026) where you state a goal and a coordinator routes work to parallel threads - each thread a Claude Code cloud session on its own branch - backed by shared memory and a library. Covers choosing the goal and repos/context, configuring environment, connectors, plugins, instructions, and model/effort for coordinator and workers, briefing the coordinator like a chief of staff, steering threads (including from a phone), handling merge order and conflicts, tuning check-in frequency and update detail, and watching project usage. Use when long-running, multi-part work (latency reduction, multi-repo migrations, endpoint retirement) would otherwise require you to split tasks, hand off context, and merge results by hand.
- ▌ Nonprofit AI Adoption · uygnoey bundleSet up and use Claude at a nonprofit following the Claude for Nonprofits plugin announced October 26, 2026 - map where the organization keeps its information, check uses against the organization's AI policy, then apply the plugin's three focus areas (extend your team, set up Claude responsibly, fund your mission) to marketing, hiring, board work, funder research, go/no-go decisions on RFPs, proposals, reports, and donor stewardship. Use when a nonprofit admin or staff member is starting with Claude on a Team or Enterprise nonprofit plan, or wants to know which plugin workflow fits a fundraising or operations task.
- ▌ Code Review Rollout · uygnoey bundleRoll out Claude Code's Code Review across an organization's repositories — enable it, install the GitHub App, pick which repositories to review, and assign each one a Quick, Standard, or Deep review depth. Use when an admin on a Claude Team or Enterprise plan is turning Code Review on, adding repositories, or deciding how deep reviews should run for a given codebase.
- ▌ Si Partner AI Rollout · uygnoey bundleStructure a global systems integrator's Claude practice along the three tracks HCLTech announced when it joined the Claude Partner Network as a Global Premier partner — (1) a dedicated business unit and Center of Excellence with forward-deployed engineers who build secure sandboxes for clients to test Claude on their own systems and data, train client workforces, and build controls for running agents at scale; (2) embedding Claude in the integrator's own software products as an ISV partner; and (3) running Claude and Claude Code inside a managed-services platform with policy guardrails, exception handling, and human review before anything touches client production systems. Use when planning or reviewing how a services firm, ISV, or regulated-industry IT organization (banking, healthcare, telecom, manufacturing) brings Claude to clients with controls suited to high-stakes environments.
- ▌ Agent Otel Tracing · uygnoey bundleSet up and use OpenTelemetry (OTel) traces from Claude Managed Agents, the Messages API, and Claude SDKs on the Claude Platform (public beta). Each agent turn, model call, tool use, and MCP server call arrives in any OTLP-compatible observability tool as a standard trace with latency, tokens, and IDs, following the OpenTelemetry GenAI semantic conventions. Covers where traces come from (Console destination for Managed Agents and the Messages API; automatic export from SDKs), picking a tool for debugging, eval/quality improvement, or org-wide usage tracking, and using traces to widen an agent's access as it proves reliable. Use when adding observability to Claude-built agents, debugging slow or failed agent steps, turning production traces into evals, or tracking token usage by team.
- ▌ Legal Artifact Workflows · uygnoey bundleRun legal work in Claude with the ready-to-use legal artifacts announced in October 2026 for Claude for Legal - cite check, jurisdiction survey matrix, matter board, playbook deviation board, regulatory horizon scan, and tabular review, plus partner artifacts (Goodwin, Harvey, Foley & Lardner, PwC) and access-to-justice artifacts. Covers opening an artifact from the legal gallery, tailoring it to a team, wiring in playbooks and research sources through legal connectors instead of manual uploads, sharing it with colleagues, clients, or partners under privacy and ethical-wall settings, and keeping a lawyer's judgment on top of Claude's sourced first pass. Use when a legal team wants a tracker, dashboard, or review table it can reuse and share, or is planning a Claude for Legal rollout.
- ▌ Revenue Org AI Rollout · uygnoey bundlePlan and run an organization-wide Claude rollout for a sales or revenue team, following Anthropic's revenue-leader guide. Covers the AI maturity ladder (individual gains → team and department workflows → plugins or managed agents pointed at KPIs/OKRs), pre-pilot decisions (RevOps as owner, connected systems, IT and security review, one activity plus one revenue metric with a baseline, spend limits by org/group/user, pilot cohort of two or three teams), the Sales plugin's commands and skills, the three-phase rollout (setup, pilot, scale in waves) with a champion ratio of one per 25-50 users, ROI readouts in three groups (efficiency, expansion, new capabilities) using pilot-vs-non-pilot comparison, and the three pitfalls that stall rollouts. Use when a revenue leader, RevOps owner, or sales manager is preparing, running, or measuring a Claude program.
- ▌ Enterprise Adoption Insights · uygnoey bundleRun the Claude Enterprise adoption loop announced in October 2026 - smart reports to see what each team does with Claude and what it costs, in-app surveys to capture members' own view of the value, and pooled group budgets to give each team one shared monthly budget. Use when an Enterprise admin or team lead wants to find high-value work, spread it as skills and repo rules, gather feedback, or fund heavy users without everyone hitting individual limits early in the month.
- ▌ Delegated Agent Safeguards · uygnoey bundleDesign or review the safeguards for an always-on agent that acts on a user's behalf in a cloud computer with connected apps and a browser - approval gates for sensitive actions, an out-of-model intent check, hard off-limits actions, outgoing-message sensitive-data checks with tiered sharing rules, task and memory visibility, credential and browser isolation from the model, and an evaluation/red-team plan. Based on how Anthropic described securing the Claude Duet research preview (October 23, 2026). Use when building, threat-modeling, or auditing a personal or delegated agent that can send messages, buy things, delete data, or sign in to sites.
- ▌ Pilot To Production Readiness · uygnoey bundleRun an enterprise AI pilot as a production-readiness exercise using the Anthropic and Accenture blueprint "Deploying AI from pilot to production" (September 2026). Walks seven considerations in order - pre-pilot (strategic objectives/ROI/ownership, data readiness and integration, infrastructure and architecture, security/compliance/trust), pilot to production (organizational readiness), and in production (governance and risk management, scale and evolution) - each closing with cross-functional "work out" questions and "assign" ownership decisions for the CIO or business leader. Includes a four-part use-case definition (user, task, output, measurable quality threshold), a lightweight total cost of ownership model, a four-tier oversight model (automated, sampled, reviewed, advisory), a manual-checkpoint audit, and error-mode analysis before scaling. Use when planning or reviewing an AI pilot, deciding go/no-go for production, assigning ownership, or designing human oversight for AI outputs.
- ▌ Phi Safe Slack Agent Rollout · uygnoey bundlePlan and run a rollout of Claude Tag (beta, Claude in Slack) inside a healthcare organization so it never touches protected health information (PHI) - modeled on the September 2026 article "How healthcare organizations use Claude Tag" with examples from Insight Health, Tennr, and Medallion. Covers admin controls (off by default, channel allowlist, DMs disabled, per-channel connector scoping, access bundles that exclude the EHR and clinical systems), pairing with a separate BAA-covered agent that masks PHI, choosing PHI-free workstreams (incident response, internal tooling, payer-rule expertise), per-channel autonomy levels, and a start-small two-week pilot. Use when a healthcare or other regulated team wants an AI teammate in Slack but must keep regulated data out of it.
- ▌ Government Agency Rollout · uygnoey bundlePlan the rollout of Claude for Government (generally available to US federal and state agencies in a FedRAMP High authorized environment) — usage-based billing with a hard not-to-exceed cap, department and sub-agency administration, SSO and SCIM seat tiers, layered configuration defaults, audit logs and ATO documentation, MDM deployment, and the early-access Claude Code CLI and Claude for Microsoft 365. Use when an agency administrator, security team, authorizing official, or procurement officer is preparing to adopt or configure Claude for Government.
- ▌ Google Workspace Editing · uygnoey bundleWork on Google Docs, Sheets, and Slides with Claude, using either the Claude for Google Workspace add-on (a sidebar beside an open file that reads the file, sees the selection, and edits in place) or the Google Docs/Sheets/Slides connectors (create or edit Google files from a Claude chat). Covers choosing the surface, picking an edit permission mode ("Ask before edits" vs "Accept all edits"), and task patterns for Docs (in-place edits, suggestion cards), Sheets (formulas, pivot tables, native charts, tabs, Python-based cleaning and joins), and Slides (layout/theme-based slides plus a layout check). Use when a user wants Claude to draft, edit, analyze, or build content inside Google Workspace files.
- ▌ Google Cloud Offering Selector · uygnoey bundleDecide between the two ways to run Claude on Google Cloud - Claude Models on Google Cloud (Anthropic managed), generally available since October 22, 2026, and Claude Models on Google Cloud (Google managed) - and plan onboarding for the chosen path. Anthropic managed gives full Claude Platform feature parity with day-zero betas (Managed Agents, code execution, computer use, agent skills, advisor strategy, Files API, citations, web search/fetch, MCP connector, prompt caching, batch processing, Fast Mode on Opus 5.5, Claude Console) behind Google Cloud IAM, one Google Cloud invoice, and commitment drawdown, with Anthropic as data processor and US/Global endpoints. Google managed fits when Google must be the data processor, strict regional data residency applies, or native integrations like Model Armor or Agent Engine are needed. Use when a Google Cloud customer asks which Claude offering to adopt, wants Claude features unavailable on their current path, or is migrating billing and access into Google Cloud.
- ▌ Small Business First Workflow · uygnoey bundleHelp a small business owner or operator (often non-technical, 5-50 employees) get a first real workflow running in Claude, following the lessons Anthropic reported in September 2026 from workshops with more than 1,000 owners - pick a run-the-business or grow-the-business task (reporting is the most common), build it in Claude Cowork with skills, connectors, or the Claude for Small Business plugin, treat Claude like a new employee whose work is checked, keep a human in the loop for review and sending, and log the workflow in a registry with PII handling noted. Use when an owner asks where to start with AI, wants to automate a report or operational task, or worries about accuracy and data security.
- ▌ Slack Personal Connectors · uygnoey bundleDecide when Claude Tag (beta, Claude in a Slack channel) should use a member's personal connectors versus the channel's admin-attached shared connectors, and configure review or auto posting. Covers how personal connectors work (connected at claude.ai/customize, used only for that person's requests, permission asked on first use, audited under the person's own account, disconnectable any time), mixing personal and shared connectors in one request, review mode versus auto mode, the limit that personal connectors never run unattended (scheduled routines and self-initiated actions use shared connectors), and admin governance options (shared tools under an agent identity, personal-only channels relying on existing role-based access, or tool-by-tool decisions; Enterprise admins can require review for everyone). Rolling out on Team plans, Enterprise to follow. Use when setting up a Slack channel with Claude Tag, choosing a connector strategy, or explaining who can see what.
- ▌ AI Insider Threat Program · uygnoey bundleDesign an "insider threat program for AI" — monitoring, real-time detection, and response for AI agents modeled on the insider threat programs companies already run for employees. Covers layering your own agent-harness telemetry (e.g. Uber's open-source AI Detection and Response, ADR) with a model provider's classifier signals (Enterprise Frontier Safeguards) and an LLM gateway perimeter, intervening close to the moment a prompt is entered or a tool is called, and a continuous loop that discovers new model behaviors and builds safeguards against them. Also covers feeding long-horizon context (code, design docs, ticket history) into security reviews and threat models. Based on Uber CISO Philip Martin. Use when planning AI security architecture, agent monitoring, or making the case for monitoring frontier-model deployments.
- ▌ Protect Cached Reads · uygnoey bundleKeep long, context-heavy Claude Code sessions cost-efficient by protecting prompt-cache hits — check how much usage is cached reads with /usage, pick the model at session start, compact before stepping away, and set the one-hour cache lifetime on API keys or cloud providers. Use when token-billed Claude Code costs are rising, when sessions run long or unattended, or when deciding which work to point Claude Opus 5.5 at.
- ▌ Continuous Security Scanning · uygnoey bundleSet up and operate continuous vulnerability scanning with Claude Security (public beta for Claude Enterprise). Covers enabling it in the admin console and assigning role-based access, creating a project that spans one or more repositories (Claude GitHub app incl. GitHub Enterprise Server, or an access token for GitLab and other Git providers) with a branch and scope per repo, the baseline scan followed by risk-based rescans of changes, writing the project knowledge base and reviewing the generated threat model so scans flag what matters and skip what the environment already handles, triaging findings by severity and attack path, automatic fixed/reopened tracking, and wiring findings into team workflows through the API, personal access tokens, webhooks, and the MCP connector in Claude.ai and Claude Code. Use when a security or platform team wants scanning that keeps up with code changes instead of one-off scans.
- ▌ Code Modernization Readiness · uygnoey bundlePrepare an organization for an agent-driven code modernization of a critical or regulated system — define the target (uplift, transform, or reimagine), write the certificate of automatically checkable conditions every change must meet, agree a tiered promotion policy for human review, stage prerequisites, then build, pilot, and scale a Claude Code dynamic workflow. Use when scoping a legacy rewrite or version upgrade, when review and change management (not code writing) are the bottleneck, or when estimating token cost for a large modernization.
- ▌ Isolate Agent Execution · uygnoey bundleDesign a secure execution setup for AI agents by splitting the "brain" (harness, session state, credentials) from the "hands" (a sandbox used only for execution), running tool calls in a self-hosted sandbox inside your own perimeter, and governing that sandbox with a default-deny policy runtime such as NVIDIA OpenShell that is tightened from its decision log toward least access. Use when building or reviewing agents that clone repos, read websites, parse tool output, or install packages, especially on Claude Managed Agents or in regulated environments.
- ▌ API Cost Performance Tuning · uygnoey bundleCut Claude API cost while holding or improving quality using the three levers from the Claude Platform best-practices article (September 2026) - maximize prompt cache hits, remove legacy prompting anti-patterns when moving to frontier models, and calibrate effort to the task - plus the claude-api skill commands that automate them (/claude-api prompt-audit, /claude-api cost-optimize, /claude-api hillclimb). Use when an API application's spend is high, cache hit rates are low, a prompt was written for an older model, or you need to choose model and effort level against an eval.
- ▌ Front Office Research Rollout · uygnoey bundlePlan and run a front-office AI rollout for a research-driven investment firm, following the pattern T. Rowe Price described in September 2026 - Claude and Claude Cowork for research synthesis and knowledge-intensive multi-step processes, Claude Code for building investment tools, business-owned applications under responsible-use standards, and human judgment kept accountable for decisions. Use when an asset manager or similar research organization wants to start AI adoption with portfolio managers and analysts, classify candidate use cases, or set up ownership and governance for them.
- ▌ Exploit Chain Validation · uygnoey bundleRun a defender workflow for model-assisted vulnerability review at portfolio scale, following how Comcast and Booz Allen used Claude Mythos Preview through Project Glasswing — hunt for exploit chains where individually correct components combine into an attack path, validate each finding against the running system, trace root cause across programs and languages, confirm whether it was exploited, draft fixes, assemble evidence for system owners, deduplicate systemic patterns, and route to owners. Use when a security team is reviewing many systems with Claude and validation, not discovery, is the bottleneck.
- ▌ Startup Program Readiness · uygnoey bundleHelp a founder check whether their startup fits the expanded Claude Startups program (founded in the last five years or funded in the last two), understand what it offers (a free year of Claude Team with up to five Premium seats for companies new to Team, a $1,000 one-time API credit, up to $45,000 in partner discounts via the Claude Startup Stack, office hours, Claude Marketplace listing help, events), and plan how to use those benefits. Use when a founder asks about Claude startup credits or the Startups program. Always tell them to confirm current status first — the cohort announced on 2026-10-06 was full as of 2026-10-13.
- ▌ Scheduled Agent Fleet · uygnoey bundleDesign, prototype, and operate a fleet of single-purpose scheduled agents for a repetitive daily workflow — one job per agent, one shared memory, scoped credentials, and a model and effort level matched to each job. Use when a recurring workflow (lead routing, triage, reply drafting, CRM updates, end-of-day follow-ups) should be split across several agents that run on a schedule, or when agents prototyped as skills plus scheduled tasks have outgrown ad-hoc editing and need one definition per agent with a versioned history.
- ▌ Coding Agent Spend Roi · uygnoey bundlePlan, review, or report on Claude Code spend and return on investment for an organization running Claude on Google Cloud, using the control -> track -> prove framework. Control sets guardrails that guide rather than restrict (Claude apps gateway spend caps per user, IdP group, or org by day, week, or month; 75%/95% in-CLI warnings and 429 at the cap; per-group model defaults and allowlists; the admin API), plus Google Cloud billing, budget alerts, spend caps, and quotas. Track uses Claude Code OpenTelemetry metrics and events (by user, team, model, skill, plugin, agent, MCP server), OTEL_RESOURCE_ATTRIBUTES cost-center tags, Cloud Monitoring dashboards, and Cloud Billing exports to BigQuery. Prove maps usage to adoption, satisfaction, and impact metrics (time, effort, volume) around a defined unit of work such as a labeled, merged PR. Use when rolling out Claude Code on Google Cloud, setting budgets or model policy, building usage dashboards, or preparing an ROI case for stakeholders.
- ▌ Fold Subagents Into Agent Loop · uygnoey bundleDecide whether each subagent in an existing multi-agent system (orchestrator plus specialists, routers, planners, domain agents) should be folded into a single main agent loop that uses progressively disclosed skills and tool search, and migrate one subagent at a time behind a task-completion eval. Keeps a subagent only when the single loop shows measurable context-bloat degradation, when work needs parallel throwaway context windows, or when one agent cannot pick tools or holds conflicting instructions. Covers failure modes of org-chart architectures (context loss, subagent sprawl, the Frankenprompt, localized evals, latency), the target single-loop design, merging coupled specialists, calling subagents as tools as an interim step, and how teams contribute skills, MCP tools, and knowledge bases under one central owner. Use when reviewing or simplifying an agent architecture, planning a subagent migration, or when a multi-agent agent is slow, inconsistent, or hard to own.
- ▌ Plan Code Mods · uygnoey bundlePlan, scope, and safely adopt Claude Code mods — small TypeScript functions, shipped inside plugins, that hook Claude Code events (tool calls, permission requests, screen drawing) to run before, after, instead of, or around them. Covers choosing a mod versus a hook, event phases, load-order stacking, UI changes in the terminal and desktop app, replacing built-in features such as /diff, team and enterprise controls (marketplace allow/block lists, managed settings, the sec-default mod), and trust rules for installing mods. Use when someone wants to customize Claude Code behavior or UI, write or install a mod, or govern mods across a team.
- ▌ Proactive Agent Delegation · uygnoey bundlePlan how to hand work to Claude Duet, the proactive agent app for professionals and small business owners (research preview, US Pro and Max subscribers via waitlist; iOS, Android, and claude.ai). Covers the three modes (morning brief, all-day monitoring of connected apps, delegated tasks from scheduling to dashboards, forecasts, prototypes, and web apps), what carries over from the Claude account (memory, skills, connected apps), choosing check-in topics and frequency, the approval gate before sending, paying, signing, or deleting, actions that stay off-limits even with approval, purchases through Link virtual cards or Shop Pay, and the preview's known limits and usage cost. Use when someone wants to set up Duet, decide what to delegate to it, explain its safety model, or set expectations about what it cannot do yet.
- ▌ Crm Seller Workflows · uygnoey bundleRun an account executive's daily work in Claude with the Salesforce in Claude plugin (beta, September 2026) - morning brief, call prep, deal review and close plan, post-meeting CRM updates, and pipeline/forecast dashboards - pulling context from Salesforce, email, call recordings, and Slack and writing changes back to Salesforce only after the seller approves. Covers admin rollout (Salesforce MCP, AgentExchange access, org-wide connection, group targeting) and the permission model. Use when a seller or sales leader wants Claude to assemble account context or keep Salesforce updated, or when an admin is planning the rollout.
- ▌ Unified App Task Handoff · uygnoey bundlePlan and hand over work in the unified Claude app, where (from September 16, 2026) Claude Cowork and chat merged into one Claude. Covers framing a bigger task inside an ordinary conversation so Claude can keep working after the laptop closes, producing Claude Docs, Claude Slides, and Claude Design outputs from the same conversation (each at one shareable link; edit directly, comment for Claude, present, or download as PowerPoint or PDF), choosing the check-in mode (ask before acting by default, or keep working and check in only when needed), scheduling recurring work such as a Monday report, and what carries over for Cowork users. Includes rollout facts (Pro and Max first; Team and Free later; Enterprise admins get at least 30 days notice; Docs, Slides, Design in conversations are beta on paid plans and admin-controlled for Enterprise). Use when someone asks where a task should go in Claude, how to hand over a report or deck, or what changed after the Cowork and chat merge.
- ▌ Marketplace Adoption Paths · uygnoey bundleRoute a team or a vendor to the right path in the Claude Marketplace, which (from September 23, 2026) brings plugins and connectors, Claude-powered agents and products, and service partners together in one place. For customers it covers adding connectors and plugins (2,000+ including Atlassian, Google, Microsoft, Notion, Salesforce), buying partner agents and products with part of committed Anthropic spend (for example CrowdStrike, Cursor, Harvey, Legora, Lovable, Snowflake), and engaging Claude Partner Network consultancies and system integrators (for example Accenture, BCG, Deloitte). For builders it covers the three listing paths - build a connector or plugin with MCP and Agent Skills, apply to list a Claude-powered agent or product, or join the Claude Partner Network as a services firm. Use when someone asks how to extend Claude through the marketplace, how to spend committed spend on partner software, or how to get their own offering listed.
- ▌ Live Dashboards And Motion · uygnoey bundleUse Claude Dashboards (beta, paid plans) to turn a plain-language question about company data into a dashboard that stays current, and Claude Motion (beta, Team and Enterprise) to turn a report, chart, or walkthrough into an editable, code-built animation exported as MP4. Covers connecting a data platform or CRM connector, checking the query and refresh time behind each chart, handing a dashboard off to a BI tool for deeper analysis, and building, editing, and exporting a Motion explainer. Use when a user wants a live dashboard from their data or an animated explainer instead of a static slide.
- ▌ Plugin Directory Submission · uygnoey bundlePlan, package, and submit a third-party extension to the Claude directory as a plugin. Covers what a plugin can bundle (MCP connectors, Agent Skills, or both; in Claude Code also LSPs, commands, hooks, and agents), the two submission paths in the developer portal (a single remote MCP connector, or a GitHub-hosted plugin bundle), automatic validation and safety scanning, reviewing status and recommended changes, choosing when to publish, reading install and discovery analytics, and the MCP 2.0 extensions (MCP Apps for interactive UI, Enterprise Managed Auth for zero-touch OAuth). Use when a developer wants to ship a connector or skill to Claude users, decide between a connector listing and a plugin bundle, or prepare for directory review.
- ▌ Enterprise Unified Rollout · uygnoey bundlePlan and run an organization-wide rollout of the unified chat + Cowork experience on Claude Enterprise (generally available October 23, 2026) - confirm eligibility, review the security posture, and set cost controls (efficient mode, model entitlements, spend limits, pooled group budgets) before enabling it in Organization settings. Use when an Enterprise admin, IT, security, or finance owner must decide how and for whom to turn on the unified experience, or must explain how cloud task work is secured and paid for.
- ▌ Long Running Agent Development · uygnoey bundleBuild agents whose sessions run for tens of minutes to hours, using Outtake's four-stage process and four hard-won learnings from their Recon Agent. Use when context compaction and behavioral drift are real rather than theoretical, when an agent keeps ignoring a system-prompt instruction no matter how it is worded, when deciding whether to stay in Claude Code or graduate to the Claude Agent SDK, when choosing between purpose-built tools and a filesystem plus bash, when manual transcript review no longer scales, or when the agent must operate in an environment that is actively trying to hijack it.
- ▌ Microsoft 365 Copilot Model Choice · uygnoeyHelp an organization map Microsoft 365 Copilot entry points (Researcher, Copilot Studio) to appropriate Claude model choices mentioned in the announcement.
- ▌ Time Bounded Extraction Context Engineering · uygnoey bundleBuild reliable, reviewable extraction prompts by assembling the right domain evidence and explicit time anchors (e.g., “most recent value before procedure start time”), and by using granular evaluation to locate failures in prompts vs context vs retrieval.
- ▌ Closing The Patch Gap · uygnoeyPrioritizes and operationalizes software patching for an environment where AI models quickly reverse patches into exploits. Use when planning patch policy, reducing time-to-patch on internet-exposed systems, triaging large CVE backlogs with KEV and EPSS, or rolling out automated patch deployment.
- ▌ Verification Loop Builder · uygnoey bundleTurn the manual checks you repeat after every change into Claude Code skills, so Claude closes its own feedback loop instead of waiting for a human to remember the check. Use when you keep making the same small corrections after Claude implements a feature; when a project-specific rule ("reject any migration that drops a column without a backfill step") is real but no generic linter catches it; when deciding whether a check should run standalone, embedded in the producing skill, chained after another skill, or on every PR; or when adding verification to a skill you cannot edit. Covers the built-in loops to try first (/verify, toolchain signals, Code Review, GitHub Actions, spec validation, rubrics in Claude Managed Agents), the minimal SKILL.md shape for a verification skill, the four deployment patterns with their costs, and the six-step creation process.
- ▌ Healthcare Agent Implementation · uygnoey bundleA practical framework for implementing AI agents in healthcare and life sciences, focusing on interoperability, latency, compliance, and maintaining human clinical authority.
- ▌ Managed Agents Private Execution · uygnoey bundleUse this skill to decide whether to run Claude Managed Agents with self-hosted sandboxes and/or MCP tunnels, and to collect the minimum deployment details and security considerations for each option.
- ▌ Agent Workflow Patterns · uygnoey bundleHelps choose and apply common AI agent workflow patterns (sequential, parallel, evaluator–optimizer). Use when designing multi-step agent systems, deciding whether to parallelize subtasks, or improving output quality with iterative evaluation.
- ▌
- ▌ Dynamic Workflows Harness · uygnoey bundlePatterns and reusable prompt templates for creating “dynamic workflows” in Claude Code to orchestrate multi-agent harnesses for complex, high-value tasks.
- ▌ Government Deployment Planning · uygnoey bundlePlan a government or regulated-environment rollout of Claude Code and Claude Cowork through Claude for Government Desktop — authorization boundary, where data is stored and processed, SCIM-based delegated administration, spend caps, audit logging, and MDM distribution. Use when preparing an ATO package or security review, when a department needs to allocate seats and limits across sub-agencies, when finance needs consumption-based spend bounded against appropriated funds, or when answering IG and auditor questions about usage without exposing sensitive material.
- ▌ School District Rollout · uygnoey bundleMove a K-12 school or district from individually verified teacher accounts to a centrally managed Claude for Teachers account — verifying the district, accepting the K-12 terms, connecting the email domain and single sign-on, and understanding what changes for teachers already using it. Use when planning a district-wide rollout, when checking eligibility and the free-access window, when confirming FERPA posture and student data handling before approving a deployment, or when deciding which teaching capabilities to introduce first.
- ▌ Multi Agent Decision Framework · uygnoey bundleDecide when to use multi-agent architectures (context protection, parallelization, specialization), how to decompose work by context boundaries, and how to add a verification subagent pattern with explicit pass/fail criteria.
- ▌ Business Admin Controls Overview · uygnoey bundleSummarize the Claude blog announcement about Claude Code availability via premium seats and the new admin controls, analytics, and Compliance API for Team/Enterprise plans.
- ▌
- ▌ Opus 4 7 Code Best Practices · uygnoeyApplies the Opus 4.7 prompting and session-tuning guidance from "Best practices for using Claude Opus 4.7 with Claude Code". Use when opening a Claude Code session with Opus 4.7, choosing an effort level (high / xhigh / max), batching questions, or deciding whether to nudge toward more tool use and parallel subagents.
- ▌ Writing Quality Vuln Reports · uygnoeyWrites upstream-quality vulnerability reports when a model has assisted in finding the bug. Use when preparing to send a finding to an open-source maintainer, vendor, or internal security team, especially if AI was involved in discovery or drafting.
- ▌ Model Class Selection · uygnoey bundleChoose which Claude model class — Mythos/Fable, Opus, Sonnet, or Haiku — to run a workload on, and use effort level to dial in the quality/speed/cost balance. Use when picking a model for a new production workload, when revisiting a model choice after evals show a gap, when a workload is latency- or cost-sensitive enough that a lower class is worth testing, or when deciding whether to pair a cheaper worker model with a more capable advisor. Default recommendation is to start with the most intelligent generally available model and tune down, because cost-per-task is often lower on more capable models even when price-per-token is higher.
- ▌ Structured Outputs Reliability Guide · uygnoey bundleChecklist and guidance for using structured outputs (schema/tool-conformant responses) to reduce parsing errors and tool-call failures in production.
- ▌ Hackathon Builder Patterns · uygnoey bundleReusable planning and iteration patterns drawn from multiple winning projects in Claude’s “Built with Opus 4.7” hackathon.
- ▌ Secure AI Native Sdlc · uygnoey bundleDesign and operate security controls for a software development lifecycle where agents author most of the code. Use when the security review queue has become the bottleneck on shipping, when deciding where human gates belong once the build step takes hours instead of months, when setting agent identity and permission boundaries (including agent-to-agent access), when introducing AI reviewers that must earn trust before they can block or approve, or when a security program needs an audit trail for decisions made by agents rather than people. Covers the Plan, Code, Test/CI, Deploy/CD, Monitor, and Governance stages with an enduring principle for each.
- ▌ Cobol Modernization Planning Playbook · uygnoey bundleA structured workflow for using AI assistance (e.g., Claude Code) to map, plan, test, and execute incremental COBOL modernization with human review and stepwise validation.
- ▌ Agentic Analytics Warehouse · uygnoey bundleUse this skill when the user asks for business analytics that requires querying a data warehouse (SQL and/or governed semantic layer), and the work must be reliable under entity ambiguity, freshness/staleness risk, and retrieval gaps.
- ▌ Spec Driven Agent Verification · uygnoey bundleShift an agentic codebase from "agents emit code" to "agents emit verifiable specifications." Use when agents generate code faster than the team can review it and verification has become the real bottleneck, when control logic is scattered across routes, database constraints, service code, background jobs, and docs so no one can state the system's state machine, when agents need to change operational behavior safely without a full CI cycle, or when designing an autonomous build-and-operate loop and deciding what humans must still coordinate. Based on Datadog's Temper, a "universal machine tool" that verifies agent-emitted specs through four independent layers before anything runs.
- ▌ Skills And MCP Architecture Playbook · uygnoeyGuidance for designing workflows where MCP servers provide tool access and skills encode procedural expertise, sequencing, and output standards.
- ▌ Code Case Study To Notes · uygnoey bundleTurn a Claude Code project case study into reusable implementation notes (problem, approach, agents, tests, constraints, and takeaways) without inventing details.
- ▌ MCP Production Integration Patterns · uygnoey bundlePatterns for connecting agents to production systems, including when to use direct APIs vs CLIs vs MCP, and how to design MCP servers and clients for scale, security, and context efficiency.
- ▌ Instruction Placement Framework · uygnoey bundleDecide where to place Claude Code instructions (CLAUDE.md, rules, skills, subagents, hooks, output styles, or appended system prompt) based on load timing, compaction behavior, context cost, and authority.
- ▌ Code Org Adoption Playbook · uygnoey bundleA reusable playbook for adopting Claude Code across an organization, based on practices highlighted in Brex’s case study (context files, documentation hygiene, and context-aware commands).
- ▌ Verifiable AI Trust Layer · uygnoeyDesign an auditable AI system by pairing Claude’s interpretation and reasoning with deterministic retrieval, computation, provenance, and evaluation so every claim can be traced to source.
- ▌ Skills Packaging Principles · uygnoey bundlePackage domain expertise into scalable Agent Skills using progressive disclosure (metadata, SKILL.md, and on-demand references).
- ▌ Skill Evals Maintenance Playbook · uygnoey bundleMaintain Agent Skills over time by defining evals, running benchmarks, comparing variants, and tightening skill descriptions to reduce false triggers.
- ▌ M365 Cross App Collaboration · uygnoeyCollaborate with Claude across Microsoft 365 apps (Excel, PowerPoint, Word, Outlook) while keeping one continuous conversation context and updating related artifacts.
- ▌ Onboarding Cc · uygnoeyTreat Claude Code onboarding like developer onboarding by maintaining durable, versioned project context (e.g., a root CLAUDE.md plus a small skill library and references to deeper documentation).
- ▌ Firm Wide AI Rollout · uygnoey bundlePlan and run a firm-wide AI assistant rollout the way a professional-services firm does it — pick the surface set rather than a single tool, pilot before going broad, wrap the launch in enablement and governance, and measure adoption by active use within weeks instead of seats provisioned. Use when deploying an AI assistant across an entire workforce, when a pilot needs to be turned into a general availability plan, when adoption has stalled after licences were handed out, or when a delivery organization wants its internal deployment to become the basis for client-facing work.
- ▌ Frontier Defense Partner Evaluation · uygnoey bundleEvaluate and shortlist the Claude Opus-powered defensive cybersecurity partner offerings that Anthropic publicized on May 21, 2026. Use when a security buyer or CISO is choosing among the three families of offerings — continuous offensive testing at production scale, closing the find-to-fix gap, and getting AI into production with governance — and needs a structured comparison against the partners Anthropic explicitly named (Wiz, Palo Alto Networks/Unit 42, CrowdStrike, Accenture, TrendAI, Deloitte, PwC), plus an awareness flag for BCG, Infosys, and SentinelOne.
- ▌ Native Sandboxing Basics · uygnoeyEstablish safer, more autonomous boundaries for agentic coding by applying filesystem and network isolation (a "native sandboxing" approach) so the agent can operate with fewer per-action permission prompts.
- ▌ Usage And Spend Governance · uygnoey bundleSet up visibility and cost control for an enterprise AI deployment using admin analytics, model defaults and entitlements, spend-threshold alerts, and programmatic access through the Analytics API and Admin API. Use when an org's usage and cost patterns stop looking like a chat tool's and start looking like agentic work, when finance or IT needs usage and cost data alongside the rest of cloud spend, when routine work is defaulting to the most expensive model, when users are hitting spend limits mid-task, or when someone asks what value the deployment is actually returning per team or per seat.
- ▌ Supervised Risk Analyst · uygnoey bundleBuild an analyst agent that works alongside human experts in a high-stakes domain — surfacing insights and forming opinions on exposure, while logging its reasoning, testing actions in sandboxes, and requiring expert approval before decisions count. Use when designing an AI teammate for risk, compliance, or another regulated analytical function; when outputs must be auditable back to their reasoning; when analysis depends on proprietary data plus judgment the firm will not delegate; or when the job is to explain how numbers changed since yesterday rather than answer one-off questions.
- ▌ Large Codebase Rollout · uygnoey bundleBest-practice playbook for making Claude Code reliable in large repositories by building a harness (CLAUDE.md layering, hooks, skills, plugins, MCP, LSP) and establishing ownership.
- ▌ Agent Fleet As Code · uygnoey bundleRun a fleet of production agents as code in a git repository — every agent a config file plus a prompt, every change a reviewed pull request, every improvement harvested from human feedback rather than retraining. Use when AI automations are scattered across individual laptops with no shared view of what exists or what it costs, when non-developers need to ship production agents without the dev team becoming the bottleneck, when agents need version history and an audit trail instead of an admin screen, or when you need a defensible way to decide which tasks deserve an agent at all.
- ▌ Human Agent Team Practices · uygnoey bundleSet up and run a team where people and agents work together in shared channels. Use when agents have workspace access but return shallow answers because context lives in private threads; when deciding which work an agent owns and where a human reviews; when defining roles for several specialized agents instead of one general assistant; when adoption is stalling and needs peer demonstration rather than mandate; or when leadership asks for evidence of value and only usage metrics are available.
- ▌ Agent Prompt Caching Best Practices · uygnoey bundleKeep prompt cache hit rate high in long-running agents by following the patterns Claude Code's team uses — lay out static-then-dynamic, push updates into messages instead of editing the system prompt, never change models or tools mid-session, model state transitions as tools (Plan Mode), defer tool loading instead of removing tools, and implement compaction as cache-safe forking. Treat cache breaks as incidents.
- ▌ Sales Cowork Workflows · uygnoeyTurn recurring sales leadership routines (meeting prep, weekly forecast rollups, and large-scale account scoring) into scheduled, repeatable workflows that assemble data from multiple systems and produce standardized outputs with human approval.
- ▌ Hackathon Project Verification Patterns · uygnoey bundlePatterns observed in Build Day projects for planning, verification with independent sub-agents, and cost/latency optimization under tight time constraints.
- ▌ Agent Orchestration Planning Layer · uygnoey bundleAdd a structured planning-orchestration layer before code generation to make outcomes, assumptions, edge cases, and validation criteria explicit, producing a reviewable PRD-style plan.
- ▌ Cowork Enterprise Rollout Playbook · uygnoey bundlePlan and run a Claude Cowork enterprise rollout the way Anthropic's own April 29, 2026 deployment guide recommends — pick a first business function, structure a pilot, walk the five-level maturity model, and execute a six-month champion-to-org-wide roadmap, anchored on the post's named internal teams and customer references (Thomson Reuters, Zapier, Jamf).
- ▌ Context Engineering For New Models · uygnoey bundleRightsize the context you hand a newer-generation model — system prompt, CLAUDE.md, skills, tool descriptions, and references — by removing rules the model no longer needs and replacing them with judgment, interface design, and progressive disclosure. Use when a system prompt or CLAUDE.md has grown long, when instructions contradict each other, when a skill over-constrains the model, when tool guidance is duplicated between the system prompt and the tool description, or when migrating an agent built for an older model generation.
- ▌ Desktop Enterprise Rollout · uygnoeyRoll out Claude Desktop across an organization with cloud-hosted inference and enterprise controls, based on a provider-backed deployment (AWS, Google Cloud, or Microsoft Foundry).
- ▌ Early Model Testing Playbook · uygnoeyA practical playbook for running a short pre-launch model evaluation window that blends automated evals with hands-on “vibe checks.”
- ▌ Long Horizon Agent Runs · uygnoey bundleHand an agent work that runs for hours unattended and get something you would actually keep — by stating the invariants up front, giving the agent the real debugging tools, and requiring it to name what it does not know instead of asserting the first plausible answer. Use when delegating a migration or a backlog of bugs overnight, when an agent drifts off-task on long runs, when triage answers stop being trusted, or when deciding whether a task is safe to leave running.
- ▌ Defensible AI Outputs · uygnoey bundleBuild AI output that a professional can put their name on — grounded in authoritative content, checked for citation validity before a human ever reviews it, and transparent enough to defend when the stakes are high. Use when the work product has to hold up under professional review (legal, tax, accounting, compliance), when deciding what to require of a model before trusting it with that work, or when deciding what the human in the loop is actually accountable for.
- ▌ Dogfood Model Evaluation · uygnoey bundleDecide whether a new model is actually better by putting it through a real day of work with your highest-taste engineers and asking whether the code is something they would keep, rather than trusting a benchmark score. Use when a model aces a benchmark but you are unsure it will hold up in practice, when building an anti-slop internal benchmark, when a score jump needs corroboration, or when a team keeps arguing for weeks about whether a release is an improvement.
- ▌ Inbound Reply Drafting · uygnoey bundleRun a sales inbox as an hourly drafting loop — a thin system prompt, a curated knowledge base of common questions and best answers, and a per-rep voice profile — so every thread needing a reply gets a draft the rep reads, edits, and sends. Use when inbound volume is consuming hours a day on repeated questions; when replies must stay factually anchored to approved product answers; when drafts should sound like the person sending them; or when adding lighter monitors for no-shows, prospects going dark, and new leads that need a first touch.