AI Agent Workflow Collection

Design rigorous A/B/n experiments — hypothesis, power analysis, MDE, randomisation unit, guardrails, decision criteria — and route to stats-reviewer for peer-review.

Quarterly life-admin sweep tailored to life stage — renewals, audits, reviews — so admin items don't drift into emergencies.

Build rule-based and statistical anomaly detection systems for business metrics — revenue drops, traffic spikes, churn increases, cost…

Multi-agent audit for Next.js + React + Supabase apps. Ten specialist auditors run in parallel, findings get validated, and a ranked evidence-backed report is written to .anthril/audits/ /REPORT.md. Read-only on source.

Read a plan-completion-audit report, plan the fixes, and execute them with safety gates per finding. Verifies between batches; optionally re-runs the audit to confirm closure.

Audit a domain's backlink profile via Ahrefs, Moz, or a free-tier fallback — producing a referring-domain register, anchor histogram, and a toxic-link disavow plan.

Crawl a domain or sitemap to find broken links (4xx/5xx), orphan pages, and soft-404s — with a prioritised remediation register.

Design complete Supabase/PostgreSQL data models with ERD, SQL migrations, RLS policies, indexes, and triggers for business…

Audit a PPC account (Google Ads or Meta) for performance issues with prioritised fixes.

Quasi-experimental design and analysis (diff-in-diff, synthetic control, ITS, regression discontinuity) for when randomised testing is infeasible. Routes to stats-reviewer.

Audit CI/CD pipelines (GitHub Actions, GitLab CI, CircleCI, Azure Pipelines, Jenkins, Bitbucket) for security, reliability, reproducibility, supply chain, and deploy safety. One sub-agent per workflow. Static, live, apply, and runtime modes.

Audit any CLI tool for terminal user experience — help text, command structure, error messages, output formatting, discoverability, and accessibility — produces a scored report with actionable…

Generate a comprehensive codebase profile — architecture topology, dependency graph, code quality metrics, security surface, test posture, and infrastructure snapshot. Stack-agnostic. Outputs to .anthril/codebase-profile.md.

Design cohort analysis frameworks with SQL queries and visualisation specs for retention, revenue, and…

Porter 5 Forces + game-theory primer for a specific market — equilibrium prediction, response-game tree, exit scenarios. Routes to red-team-strategist agent.

Audit one or more competitor domains across indexed footprint, content topics, top keywords, backlinks, on-page patterns, and technical signals — produces a comparative matrix vs your domain.

Audit Dockerfiles and docker-compose files for base image, user privileges, secret leaks, layer efficiency, signal handling, healthchecks, and compose safety. One sub-agent per Dockerfile. Static, live (Trivy/Grype), and apply modes.

Audit Core Web Vitals (LCP, INP, CLS) for a URL list or sitemap via PageSpeed Insights and CrUX — producing a per-URL scorecard, worst-offender summary, and root-cause remediation plan.

Map fixed vs variable costs by line, contribution-margin waterfall, and scale curve (how costs change at 2× / 5× / 10× volume).

Design a daily wellness habit stack (hydration / movement / sunlight / breathwork) anchored to existing routines via the habit-stacker pattern. Low effort, high frequency.

Auto-generate comprehensive data dictionaries from database schemas, CSV files, or API responses with column definitions, relationships, and Mermaid…

Design ETL/ELT pipeline architectures with data flow diagrams and transformation specs for Supabase and…

Profile datasets and audit data quality across six dimensions, producing prioritised cleaning…

Compare avalanche vs snowball debt payoff strategies, scan refinance options, build a payoff timeline, and produce a behavioural staircase plan for AU debt holders.