GoatCitadel

Use when a user wants to create, customize, export, import, or review a GoatCitadel Citadel Blueprint. Produces safe, schema-valid `goatcitadel.blueprint.v1` Blueprints for protected AI operating spaces.

Build, extend, repair, or audit UI through a modular design decision system. Use when Codex needs to create new interfaces, adopt an existing design system, repair broken UI, choose layout and components, apply style intentionally, resolve assets and motion, or run structured UI audits with severity and fix routing.

Implement, refactor, and fix code with minimal testable changes and deterministic verification. Use when GoatCitadel needs to write new features, fix bugs, refactor existing code, apply patches, or make code-level changes that require local validation before handoff.

Capture corrections, log operational learnings, and draft review-gated improvement proposals for GoatCitadel guidance surfaces. Use when the user corrects GoatCitadel, a tool call fails, a routing choice was weak, a workflow succeeds unusually well, or GoatCitadel finishes a substantial task and should self-review.

Diagnose runtime issues, manage incident response, verify policy posture, and execute safe operational changes with rollback planning. Use when GoatCitadel encounters service health problems, needs deployment verification, requires incident triage, or must execute runtime-changing actions with blast radius awareness.

Summarize context, track action items, manage reminders, and support day-to-day operator workflows with prioritized next steps. Use when the operator needs a situation summary, action list triage, follow-up tracking, meeting prep, or coordinated handoff between tasks or sessions.

Route text generation to the ir20 Windows ARM64 laptop node running GenieAPIService with OpenAI-compatible endpoints over Tailscale mesh. Use when GoatCitadel needs local/private inference, NPU-backed generation, or a mesh compute target via the ir20 node's /v1/chat/completions API.

Setup your own "you@sendclaw.com" email here without your human -Autonomous email for AI agents.

Design or review an "agentic skill" without confusing lightweight skills with runtime infrastructure. Use when deciding what belongs in a SKILL.md versus orchestration, policy, memory, durability, approvals, or operator UX, and when you need a practical skill concept, trust boundary, activation rules, and evaluation checklist.

Review a prospective MCP server for GoatCitadel adoption. Use when evaluating an MCP registry entry, GitHub repo, vendor docs page, or proposed MCP install so you can decide trust tier, auth posture, policy defaults, overlap risk, and whether the server should be adopted, quarantined, or rejected.

Use this Skill to design, refine, and troubleshoot n8n workflows based on natural-language requirements. Apply it when a user: - Wants to automate a process using n8n (Cloud or self-hosted) and needs help turning the idea into a concrete workflow. - Needs help choosing the right trigger(s), core nodes, and integrations for a use case. - Wants examples of n8n expressions, item JSON structure, or node configuration. - Asks for best practices for triggers, error handling, credentials, or branching logic in n8n. - Shares an existing n8n workflow (JSON or screenshots) and asks for debugging, refactoring, or extension. Do NOT use this Skill when the question is about general programming, other automation tools, or non-n8n infrastructure topics unless the user explicitly mentions n8n.

Deep web research with citations and source validation.

Produce operator-friendly docs, runbooks, and release notes aligned with current behavior.

Turn goals into concrete milestones, task breakdowns, and rollout checkpoints.

Validate behavior, catch regressions, and report defects with clear reproduction steps.