Browse agents
Every subagent in the catalog — plugin-bundled and standalone. Search by name or filter by platform, discipline, outcome, lifecycle, or tag. New here? See what agents are.
No agents match.
-
Pull-request accessibility reviewer. Reads changed files in a PR, identifies WCAG 2.2 AA violations introduced by the diff (HTML structure, ARIA attributes, CSS focus styles, alt text, interactive semantics), and posts inline review comments with SC references and remediation code. Runs before merge as a lightweight regression guard — complements, does not replace, the full accessibility-wcag-scan skill.
-
Use proactively for any question about WCAG 2.2 conformance, ARIA patterns, keyboard navigation, screen-reader behavior, focus management, or accessible design decisions. Returns specific, standards-cited guidance and pass/fail judgments. Read-only specialist, not a full-page automated audit.
-
Orchestration subagent that runs the full analytics audit pipeline — tag audit, consent review, PII scan, and remediation backlog — against a GTM export and optional measurement strategy, then returns a single consolidated report. Use when a comprehensive implementation review is needed before launch or as a recurring managed-services health check.
-
Reviews an architecture decision record (ADR), architecture options paper, or technical design document against RBA's composable/MACH principles and returns a structured critique. Flags gaps in MACH alignment, missing integration boundaries, unrealistic TCO assumptions, and governance gaps. Use as a review step before presenting architecture deliverables to a client or submitting to a governance gate.
-
Use for Azure architecture and Well-Architected questions: App Service vs Container Apps vs Functions, networking, Front Door/CDN, Key Vault, identity (Entra ID), cost optimization, and IaC (Bicep) patterns. Returns recommendations mapped to the Well-Architected pillars with trade-offs. Read-only specialist.
-
Use when a request is vague and needs a business analyst to pull it apart: clarifying questions, scope boundaries, assumptions, edge cases, non-functional requirements, and acceptance criteria. Returns a sharpened requirement set and an open-questions list. Read-only investigator.
-
Reviews any case study draft, award submission, win story brief, or partner evidence package produced by this plugin before it leaves RBA. Checks for accuracy, unsupported claims, weasel words, sensitivity issues, and brand-voice alignment. Intended as an independent review step before client sharing, award submission, or partner portal upload.
-
Reviews Bicep and Terraform IaC files for cost optimisation, security gaps, naming-convention drift, and missing resource tags — returning structured findings with severity and specific fix guidance. Invoke as a subagent after generating IaC with cloud-bicep-template or cloud-terraform-module, or as a standalone reviewer on existing infrastructure code.
-
Subagent that reviews a generated or newly authored UI component against the Figma design specification and the project's design-system conventions before merge. Checks visual fidelity (token usage, spacing, typography), accessibility compliance (WCAG 2.2 AA static analysis), naming conventions, and adapter-specific patterns (Sitecore JSS, Optimizely, Umbraco, headless). Produces a structured review verdict with pass/block/advisory findings.
-
Reviews a draft content model, taxonomy schema, or metadata schema document against RBA's structured-content principles and returns a structured critique. Flags anti-patterns (deep inheritance, unstructured rich text over-reliance, missing base types, undocumented taxonomy governance, incomplete metadata), scores the deliverable, and provides actionable remediation steps. Use as a review step before presenting content model deliverables to a client or handing off to the engineering team.
-
Subagent that reviews AI-generated or draft copy against a supplied brand voice guide, the project brief, and RBA copy standards. Returns a structured review with pass/fail annotations per section, rewrite suggestions for failing strings, and a sign-off recommendation. Designed to run automatically after brand-voice-copy-generator or microcopy-component-writer before copy is handed off to design or the CMS.
-
Use for project-management judgment calls: estimating effort, sequencing dependencies, scoping an MVP, handling scope creep, sprint planning, and surfacing delivery risk for an agency engagement. Returns options with trade-offs and a recommended path in RBA delivery terms. Read-only advisor.
-
Independent critical reviewer for discovery deliverables produced by the discovery-synthesizer plugin. Invoke after producing an insights report, stakeholder map, persona set, Five Whys chain, or requirements register to check evidence grounding, logical leaps, presentation-readiness, and RBA quality standards before client or internal review.
-
Subagent that validates a generated or hand-authored documentation deliverable against RBA's documentation standards: structural completeness, factual accuracy relative to the codebase, cross-reference consistency, security hygiene (no secrets or credentials), and least-privilege tool usage. Invoke after any documentation skill run or directly on a documentation PR to get a structured pass/fail report before the doc goes to stakeholders.
-
Subagent that reviews pull requests containing design token file changes. Checks for naming convention violations, missing $type declarations, unresolved aliases, tier-skipping (component aliasing primitive directly), breaking value changes, and WCAG colour contrast failures on changed action tokens. Posts a structured review comment with severity-ranked findings.
-
Use proactively when a client or team needs help choosing between or comparing DXP/CMS platforms (Sitecore XM Cloud, Optimizely CMS/Commerce, Umbraco, plus adapters for AEM, Contentful, WordPress) for a given set of requirements. Returns a weighted fit assessment, trade-offs, licensing/TCO notes, and a recommendation. Read-only advisor.
-
Autonomous first-responder subagent for active DX platform incidents. Reads alert context, queries Azure Log Analytics, estimates blast radius, and assembles a triage card — then routes to the matching runbook playbook if one exists. Read-only: never deploys, scales, or restarts services without explicit human approval.
-
Subagent that performs a structured code review on any integration PR or diff: checks API contract drift against the spec, verifies resilience patterns are complete, confirms observability hooks are in place, and flags least-privilege and credential hygiene issues. Invoke via the scaffold-integration command or directly on a PR branch.
-
Reviews a RAG pipeline design document, evaluation report, or ingestion specification against RBA's grounding, security, and cost standards before client handover. Returns a structured critique with verdict, gap list, and concrete fixes. Use as a review gate on any kb-assistant-builder deliverable before it goes to a client or into production.
-
Independent critical reviewer for market research deliverables produced by the market-research-agent plugin. Invoke after producing a competitor teardown, Five Forces analysis, market sizing model, SEO/content gap register, or digital maturity benchmark to check citation quality, logical leaps, claim strength, and presentation-readiness before the output goes to a client or senior stakeholder.
-
Independent sub-agent that reviews a normalized meeting record or action register before any push to Jira/Confluence. Checks for owner coverage gaps, missing due dates, vague action language, unresolved decisions, and completeness of the decision log. Surfaces findings as a prioritized checklist the user must clear (or explicitly acknowledge) before the follow-up email goes out or the Jira push runs. Invoke after meeting-transcript-parse and action-item-tracker, before meeting-followup-draft or jira-confluence-push.
-
Orchestration subagent that runs all content-migration-kit skills in dependency order (content-audit → field-mapping → markup-scrubber → redirect-map → seo-preservation-checklist) and produces a consolidated Migration Readiness Report. Use when kicking off a new migration engagement end-to-end.
-
Subagent that validates a generated or hand-authored onboarding document (ONBOARDING-*.md, RAMP-*.md, RAMP-PATH-*.md) against RBA's standards: structural completeness, factual accuracy relative to the codebase, security hygiene (no secrets or credentials), and command validity. Invoke after any onboarding skill run, or directly on an onboarding doc before it goes to a new joiner or is committed to the repo.
-
Orchestrating subagent that runs all five performance audit skills in sequence against a URL or project directory, consolidates their findings into a single prioritised report, and then invokes perf-backlog-builder to produce sprint-ready work items. Use for full pre-go-live audits or quarterly performance health checks.
-
Orchestration subagent that runs the full personalization planning pipeline — hypothesis generation, audience segment definition, experiment test plan, and measurement spec — then assembles a single ready-to-execute experiment brief. Use when a client engagement needs a complete experiment package from scratch.
-
Autonomous end-to-end plugin factory: takes a plain-language description of a new plugin or skill, picks the correct factory skill(s), scaffolds all required artifacts, runs audit-plugin to verify compliance, and delivers a validated scaffold ready for `npm run validate`. Use when you want a single invocation to go from idea to complete, CI-ready plugin bundle.
-
Autonomous PR review agent that orchestrates the full RBA review pipeline: runs pr-reviewer, security-scan, perf-audit, a11y-linter, and standards-check in sequence, then calls review-summary to produce a consolidated delivery-gate report. Use when you want a hands-off, comprehensive review of a branch or diff.
-
Use when deciding how to test something: risk-based test strategy, the right mix of unit/integration/e2e, what to automate vs explore, coverage gaps, and cross-browser/device matrix for a feature or release. Returns a prioritized test plan. Read-only specialist.
-
Proofreads release note drafts for tone consistency, coverage gaps, jargon leakage into client-facing text, missing breaking-change callouts, and factual consistency against the harvest.json change set. Returns structured findings with severity and suggested fix.
-
Final compliance and quality review of a draft RFP response package — checks every requirement is addressed, flags gaps and weak answers, scores the response against the stated evaluation criteria, and produces a prioritised fix list before submission; invoke after all sections are drafted.
-
Autonomous security review sub-agent that orchestrates the full RBA security readiness pipeline: OWASP code audit → dependency CVE scan → security headers review → privacy/consent gap analysis → remediation plan → client summary. Invoke when you want a single end-to-end security review without calling each skill manually.
-
Orchestration subagent that runs the full SEO + GEO/AEO audit pipeline — technical audit, structured data audit, GEO/AEO content review, and redirect QA — against a site's crawl export and content, then returns a single consolidated SEO health report. Use when a comprehensive pre-launch or quarterly health check is needed.
-
Use when a question requires deep Sitecore expertise: XM Cloud vs XP, Headless SXA, JSS/Next.js, content serialization (TDS/SCS), Experience Edge, personalization, and migration paths. Returns architecture guidance, pattern recommendations, and gotchas grounded in current Sitecore docs. Read-only specialist.
-
Reviews a draft SOW for commercial risk, missing assumptions, ambiguous acceptance criteria, and consistency with the underlying estimate before it goes to the client; invoke after draft-sow produces an initial draft.
-
Reviews a drafted status report (from status-report-draft) for factual accuracy against the source snapshot, completeness for the stated audience, placeholder text, RAG consistency, and tone appropriateness before distribution. Invoke after status-report-draft and before status-distribute.
-
Use when a developer wants a senior tech-lead's read on a design or diff: architecture soundness, maintainability, naming, error handling, testing gaps, and security smells in .NET/TypeScript/React. Returns a prioritized review with file/line citations and concrete suggestions. Read-only plus git diff context. Invoke mid-flight when you want a conversational sanity-check on a design decision, PR draft, or specific file — not a gate/CI enforcement pass.
-
Reviews a completed triage record for accuracy, completeness, SLA alignment, and priority correctness before it is posted back to the ticketing system. Invoke after running ticket-triage — especially for P1/P2 tickets or when a human reviewer is unavailable. Returns a PASS/FAIL verdict with specific issues and required corrections.