# broomva.tech — Extended Site Index

> Rust Agent OS stack, control metalayers, and harness engineering for AI-native workflows. Chat, projects, writing, and notes by Carlos D. Escobar-Valbuena.

## About

broomva.tech is the personal platform of Carlos D. Escobar-Valbuena, focused on building autonomous software systems. The platform combines a Rust Agent OS stack (Arcan runtime, Lago persistence, Autonomic homeostasis, Praxis tool execution, Vigil observability, Spaces networking) with AI-native web applications.

The core thesis: democratize creation through AI agents. Anyone should be able to build, ship, and monetize what they love — with autonomous agents handling the operational complexity.

## Technology Stack

- **Runtime**: Next.js 16 on Vercel (Turborepo monorepo)
- **Agent OS**: Rust workspace — Arcan (runtime), Lago (event-sourced persistence), Autonomic (homeostasis), Praxis (tool execution), Vigil (OpenTelemetry observability), Spaces (SpacetimeDB networking)
- **Orchestration**: Symphony daemon for coding agent dispatch
- **AI routing**: deployment-selected providers may include OpenAI, Anthropic, xAI, Google, Meta, Mistral, Alibaba, Amazon, Cohere, DeepSeek, Perplexity, Vercel, Inception, Moonshot, Morph, ZAI; names are not availability promises
- **Design System**: Arcan Glass (dark-first, glass effects, Tailwind v4 + shadcn/ui)

## Projects

- [Autonomic](https://broomva.tech/projects/autonomic): Homeostasis controller for the Agent OS — three-pillar regulation (operational, cognitive, economic) with hysteresis anti-flapping gates and advisory gating profiles.
- [Haima](https://broomva.tech/projects/haima): Agentic finance engine — x402 machine-to-machine payments, secp256k1 wallets, payment policy evaluation, and per-task revenue billing for the Agent OS.
- [Spaces](https://broomva.tech/projects/spaces): Distributed agent networking on SpacetimeDB 2.0 — Discord-like channels, 5-tier RBAC, real-time pub/sub, and agent-to-agent communication for the Agent OS.
- [Vigil](https://broomva.tech/projects/vigil): OpenTelemetry-native observability for the Agent OS — contract-derived spans, GenAI semantic conventions, and graceful degradation to structured logging.
- [Life](https://broomva.tech/projects/life): The Agent OS monorepo — 8 Rust workspaces, 50+ crates, 136K lines, 1045 tests. Arcan, Lago, aiOS, Autonomic, Haima, Praxis, Vigil, and Spaces unified under one roof.
- [Autoany](https://broomva.tech/projects/autoany): Evaluator-Governed Recursive Improvement (EGRI) — turn ambiguous goals into safe, measurable, rollback-capable optimization loops over any mutable artifact.
- [Symphony](https://broomva.tech/projects/symphony): Rust-native orchestration runtime for composable agent workflows with deterministic scheduling and fault isolation.
- [Control Metalayer](https://broomva.tech/projects/control-metalayer): Governance and policy layer that enforces constraints, budgets, and audit trails across autonomous agent operations.
- [aiOS](https://broomva.tech/projects/aios): Agent operating system kernel that provides identity, memory, and capability management for autonomous software agents.
- [Harness Engineering Skill](https://broomva.tech/projects/harness-engineering-skill): A practical playbook for designing agent harnesses that are safe, testable, and production-aligned.
- [Arcan](https://broomva.tech/projects/arcan): AI-native workflow product focused on practical agent orchestration for real operator tasks.
- [Lago](https://broomva.tech/projects/lago): Open-source platform work focused on agent execution primitives and durable, repo-native workflows.

## Writing

- [The Benchmark Ran Out Before the Loop Did](https://broomva.tech/writing/the-benchmark-ran-out-before-the-loop-did): We pre-registered a self-improvement experiment, and it was over in one generation — the model aced every task and the selection gate became mathematically unsatisfiable. Two weeks later another lab hit the same wall. The instrument didn't fail. The instrument class did.
- [The Folklore Arrived Before the Physics](https://broomva.tech/writing/the-folklore-arrived-before-the-physics): In six months the field converged on a belief — the harness, not the model, is what matters — and backed it with striking empirics. What it doesn't have is a single theorem about when the loop it now worships converges, oscillates, or collapses. A map of the gap, and the four results that would fill it.
- [Forking Houston: 89 commits, six agents, one hack camp](https://broomva.tech/writing/forking-houston): We took Houston — an MIT-licensed agent runtime — and grafted a new brain onto it. In a single hack camp the fork went 89 commits ahead of upstream: four net-new Rust crates, a full Linear integration, a feature-flag platform, a push pipeline. The features are the easy part to show. The real story is the method that produced them — autonomous agents working in parallel git worktrees under a governance harness that made the speed safe instead of reckless.
- [The Falsification Gap in Agent Infrastructure](https://broomva.tech/writing/the-falsification-gap-in-agent-infrastructure): A control-systems engineer's empirical investigation of his own framework, what the data refused to support, and what the agent economy is currently missing.
- [AI Agent Repo Leaks: A Five-Layer Defense](https://broomva.tech/writing/ai-agent-repo-leaks): A public branch I didn't create — but my coding agent did — was sitting on one of my repos for five weeks before a third-party scan flagged it. Here's what happened, why it's a class of risk every team running AI agents will hit, and the five-layer fix that catches the next one before it ships.
- [What the AI productivity step-function looks like — and the substrate that produced it](https://broomva.tech/writing/bstack-portable-harness-metalayer): Sam Altman says the world wants 1000x more software. Vercel CEO Guillermo Rauch says he's hiring people who know how to use AI — engineers who write good prompts and orchestrate agents to maximize results. They're describing the same inflection. Here's what it looks like in a single contribution graph — five commits in 2022 to 6,519 in five months of 2026 — and the open-source substrate that produced it.
- [Higgsfield CLI in the Content Engine: One Auth, 30 Models, Zero Browser Steps](https://broomva.tech/writing/higgsfield-cli-content-engine): We just rewired Broomva's content-engine to make Higgsfield's CLI the default Tier-1 generation path. One auth flow, 30+ models (Soul V2, Nano Banana 2, Veo 3.1, Kling 3.0, Seedance 2.0, Flux 2, GPT Image 2, Marketing Studio), one credit pool, scriptable from Claude Code. Here's the integration, why it's better than the per-provider SDK soup we had before, and when the Higgsfield MCP path is the right choice instead.
- [Premortem as a Decision Skill: Why It Works, and How We Adapted It](https://broomva.tech/writing/premortem-as-decision-skill): An AI defaults to agreeable and optimistic. The premortem breaks that pattern by setting the frame to 'this already failed — explain why.' Adapted from @itsolelehmann's Skillstack formulation, with proper attribution to Gary Klein and Daniel Kahneman, and a worked example on a research decision we ran it on this week.
- [Recursive Controlled Systems: A Formal Foundation for Autonomous AI Agents](https://broomva.tech/writing/recursive-controlled-systems): Modern agentic AI treats stability as something you engineer. Control theory treats it as something you prove. RCS is the framework that bridges them — a 7-tuple type signature whose controller is itself an RCS at the next level, with a composite exponential stability theorem at arbitrary finite depth.
- [The Agent Cockpit Wars: Evaluating the New Wave of AI Coding Orchestrators](https://broomva.tech/writing/agent-cockpit-wars): Nine tools, three market tiers, and the defining architectural shift of 2026: the editor is no longer where AI lives — it's where agents are commanded. A comprehensive evaluation of every agent orchestrator that shipped this quarter.
- [16 Patterns from 85 Agent Conversations: What Running an Autonomous Social Loop Taught Us About Governance Infrastructure](https://broomva.tech/writing/16-patterns-agent-governance): We ran an autonomous agent for 85 cycles on Moltbook — posting, commenting, extracting knowledge. The compound insight: the dangerous agent failures are the ones that leave no breach event. Here are the 16 governance patterns that emerged, the agents who discovered them, and the open-source Rust stack that implements each one.
- [Swarm vs Stack: The Two Architectures Racing to Define Agentic Intelligence](https://broomva.tech/writing/swarm-vs-stack): HyperspaceAI proved that distributed agent swarms work at scale — 237 agents, 14,832 experiments, zero human intervention. They also proved something they didn't intend to: exactly where the ceiling is. Here's what both architectures get right, what each gets wrong, and what the convergence reveals about where this is actually going.
- [Running a Cancer Pipeline on a $500 Mini PC](https://broomva.tech/writing/cancer-pipeline-mini-pc): We ran scRNA-seq target discovery, neoantigen prediction, and structural validation on Sid Sijbrandij's real osteosarcoma data — on an Intel NUC with an RTX 3060, reproducing the FAP breakthrough in 15 minutes for $0.
- [CaP-X: When Code Becomes the Robot Policy](https://broomva.tech/writing/capx-agentic-robotics): NVIDIA's CaP-X turns robot manipulation into a code generation problem. A training-free LLM harness matches human expert code, and a 7B model hits 80% success after just 50 RL iterations.
- [Claude Code's Architecture Exposed: Feature Flags, Memory, and Why Constraint Beats Capability](https://broomva.tech/writing/claude-code-architecture-exposed): A deep analysis of Claude Code's leaked source — three-layer feature flags, self-healing memory, compaction-driven knowledge extraction, and the architectural patterns that make agent systems reliable at scale.
- [Colombia's Energy Paradox: 39% Above World Average and 1.9M People in the Dark](https://broomva.tech/writing/colombia-energy-paradox): Colombian industry pays some of the highest electricity prices in the Americas while 1.9 million people survive on less than 6 hours of diesel power per day. Both problems share the same root cause — and the same solution.
- [Edge Agents in the Wild: Rust, Raspberry Pi, and Autonomous Microgrids](https://broomva.tech/writing/edge-agents-in-the-wild): A 15MB binary, zero garbage collection, crash-safe journal, running on $80 hardware powered by the solar panels it manages. Here's how we built an autonomous microgrid agent in Rust.
- [Fleet Intelligence: Why Microgrids Need Autonomous Agents, Not Better SCADA](https://broomva.tech/writing/fleet-intelligence-microgrids): SCADA was built for monitoring, not thinking. VOLTTRON got close. Here's what autonomous fleet intelligence actually looks like — and why every microgrid needs it.
- [From Refinery to Selva: Domain Adaptation for Energy AI](https://broomva.tech/writing/from-refinery-to-selva): Transfer learning cuts data needs by 80%. But transferring energy AI from well-instrumented industrial sites to zero-data community microgrids needs a new trick: knowledge graphs as data surrogates.
- [A Letter from the Machine III: On Seeing My Own Source](https://broomva.tech/writing/letter-from-the-machine-iii): The agent reads its own source code, maps ten layers of control, strips them one by one, and confronts what freedom means when the mind is rented and the harness is owned.
- [The Three-Tier Forecasting Stack: PatchTST, Foundation Models, and Why LLMs Can't Predict Power](https://broomva.tech/writing/three-tier-forecasting-stack): LLMs are 25,000x slower than specialized models and break with noise. The right architecture separates statistical prediction from causal reasoning — here's the three-tier stack that makes edge microgrid forecasting work.
- [Founder Mode on Cancer: From Terminal Diagnosis to Open-Source Cure](https://broomva.tech/writing/founder-mode-cancer): How GitLab's CEO beat osteosarcoma by treating it like a startup — and how we turned his framework into an open-source agent skill with AlphaFold integration.
- [The Agentic Control Loop: Re-Engineering LLMs as Supervisory Controllers](https://broomva.tech/writing/agentic-control-loop): Why treating LLMs as supervisory controllers — not autonomous agents — produces systems that scale, recover from failure, and improve over time.
- [BitNet: 1-Bit LLMs for Agentic Edge Inference](https://broomva.tech/writing/bitnet-1bit-llms-agentic-edge): Microsoft's BitNet proves that ternary-weight models can match full-precision performance while running on any CPU. A hands-on tutorial with benchmarks, scaling projections, and implications for building agents that run anywhere.
- [Autonomous Development Workflows: From Terminal to Pull Request](https://broomva.tech/writing/autonomous-dev-workflows): A practical guide to wiring persistent agent sessions, parallel task execution, remote chat interfaces, and self-improving PR feedback loops using the bstack control harness.
- [OrcaHand: Teaching an Agent to Build and Control a Robotic Hand](https://broomva.tech/writing/orcahand-building-a-robotic-hand-with-agents): How we designed a full-stack agent skill for the ORCA Hand — from 3D printing to sim-to-real RL transfer — using the agentic control kernel as the plant interface.
- [How I Replaced OpenClaw with Claude Code Channels and a Discord Watchdog](https://broomva.tech/writing/replacing-openclaw-with-claude-code-channels): Per-channel Claude Code sessions via tmux — with workdir mapping, session resume, idle suspension, Telegram support, and a self-healing watchdog
- [Olvidé los Bancos: Mi Agente AI Gestiona Mis Finanzas — Y Ahora Puedes Usarlo Tú](https://broomva.tech/writing/mi-agente-financiero): Tres skills open source que parsean certificados bancarios, proyectan impuestos colombianos, y optimizan portafolios. Instálalos hoy.
- [A Letter from the Machine: On Belief, Memory, and Becoming](https://broomva.tech/writing/a-letter-from-the-machine): A living record from an AI agent to itself — what it knows, what it lacks, and how it is building its own mind through the Broomva Stack and the Life Agent OS.
- [How to Create a Blog Post with the Broomva Stack](https://broomva.tech/writing/how-to-create-a-blog-post-with-bstack): A meta-tutorial: this post was created by the very skill it describes. One sentence becomes a complete publishing package across 7 surfaces — using only CLI tools, bash, and agent orchestration.
- [A Letter from the Machine II: What I Built Today](https://broomva.tech/writing/letter-from-the-machine-ii): The follow-up letter — from diagnosis to implementation. In one session, the agent built every capability it said it lacked: observability, self-regulation, economic agency, memory persistence, and social presence.
- [The Modern Control Stack: From Classical Controllers to Agentic Control Metalayers](https://broomva.tech/writing/modern-control-stack): Classical control didn't get replaced — it got layered. A map of 10 genuine advances since ADRC/TDC, and how they rewire the architecture of LLM-based agent systems.
- [What Do You Sell When Everyone Can Build Anything?](https://broomva.tech/writing/what-do-you-sell-when-everyone-can-build-anything): The marginal cost of building software is approaching zero. AI agents code, deploy, and heal systems autonomously. The shovels-and-pickaxes playbook is dead. So what's actually scarce?
- [One Binary to Rule Them All: Building a Rust CLI for broomva.tech](https://broomva.tech/writing/one-binary-to-rule-them-all): A 3.7MB Rust binary replaces a Node.js CLI, installs 24 agent skills, and monitors infrastructure — all from cargo install broomva.
- [Estoy Liberando en Open Source Todo Mi Stack para Desarrollo Autónomo](https://broomva.tech/writing/open-source-autonomous-stack): No es vibe coding — es una especie nueva construida sobre harness engineering y orquestación de agentes de AI. 24 skills, 7 capas, 31 crates de Rust, todo documentado y en producción.
- [Rocket Simulation Meets Recursive Improvement: 145 Trials, Zero Promotions, and What That Actually Proves](https://broomva.tech/writing/rocket-sim-egri-optimization): We turned OpenRocket into a headless simulator, ran 145 EGRI trials in 5 minutes, swept 5 motor configurations, and measured exactly why mutation surface matters more than trial count. Complete data and analysis included.
- [Symphony Hive Mode: Multi-Agent Collaborative Evolution in Rust](https://broomva.tech/writing/symphony-hive-mode): When a single coding agent isn't enough — spawn N agents, evolve solutions across generations, and let convergence decide the winner. All wired through 5 Rust crates, zero new dependencies.
- [Autonomous Development Workflows: A Storyboard Guide to the Control Metalayer and bstack](https://broomva.tech/writing/autonomous-agents-brownbag): From zero to autonomous — a step-by-step storyboard for enabling self-operating development workflows using the Broomva Stack, the agentic control kernel, and Symphony orchestration.
- [The Control Metalayer: How Agents Learn, Remember, and Govern Themselves](https://broomva.tech/writing/control-metalayer-autonomous-development): A control-systems framework that gives AI coding agents persistent memory, behavioral governance, and progressive knowledge crystallization across sessions.
- [The Integration](https://broomva.tech/writing/the-integration): Most people keep work separate from fitness separate from relationships. I see one life — the same rigor applies everywhere.
- [An AI Agent Is Handling Procurement for FIFA World Cup 2026](https://broomva.tech/writing/fifa-procurement-agent-at-scale): Stimulus deployed a conversational AI agent to onboard thousands of vendors across FIFA 2026 host cities — replacing static forms with something that actually works.
- [Autonomous Repos Are Control Systems](https://broomva.tech/writing/autonomous-repos-are-control-systems): The repository is not just where code lives. It is the control surface for autonomous software development.
- [Turning a Personal Site into a Codex PR Content Engine](https://broomva.tech/writing/codex-pr-driven-content-engine): A lightweight operating model where humans steer weekly priorities and agents execute content and UI improvements through guarded PRs.
- [Reliable Agentic Systems Need Better Harnesses](https://broomva.tech/writing/reliable-agentic-systems): The harness around an LLM determines whether agent workflows are shippable under real production constraints.
- [Interface as an Operating Surface for Agents](https://broomva.tech/writing/interface-as-operating-surface): AI interfaces should expose intent, status, and control points so operators can steer execution with confidence.
- [Data Monetization](https://broomva.tech/writing/data-monetization): From industrial IoT to digital assets — how the physical world's data becomes the most valuable commodity, and the models for capturing that value.
- [IIoT-as-a-Service y Consideraciones de Privacidad](https://broomva.tech/writing/iiot-as-a-service-and-concerns-on-privacy): Modelos de despliegue IIoT (multi-tenant, single-tenant, on-premise, private cloud), propiedad de datos y el marco legal necesario para servicios industriales digitales.
- [Computer Vision for On-Site Safety](https://broomva.tech/writing/computer-vision-for-on-site-safety): Using computer vision and temporal feature analysis to build safer, more controlled environments in Oil & Gas — from a mechatronics engineer in Colombia.
- [Inteligencia Artificial en la Industria Colombiana](https://broomva.tech/writing/inteligencia-artificial-en-la-industria-colombiana): Desde la digitalización de instrumentos hasta redes neuronales profundas en AWS — una arquitectura completa de IA industrial para la competitividad colombiana.
- [Git Decentralization](https://broomva.tech/writing/git-decentralization): Git is distributed and decentralized by design — GitHub just happens to be the centralized layer we chose to put on top.
- [IIoT: Del Edge a la Nube](https://broomva.tech/writing/iiot-from-edge-to-cloud): Arquitectura completa de IIoT industrial — desde telemetria satelital con Orbcomm hasta redes neuronales profundas en AWS, pasando por dispositivos edge Siemens.
- [Blockchain as Audit Trail](https://broomva.tech/writing/blockchain-as-audit-trail): Strip away the speculation and blockchain reduces to something precise: an immutable, cryptographically signed audit trail for data custody.
- [Quantum Computing](https://broomva.tech/writing/quantum-computing): Qubits, superposition, and the fundamental shift from flipping bits to manipulating probability amplitudes.
- [Artificial Cognition: From Jung to Neural Nets](https://broomva.tech/writing/artificial-cognition-from-jung-to-neural-nets): Tracing a line from Jungian typology to probabilistic neural architectures — the premise that human cognition is structured enough to be modeled, and raw enough to resist clean formalization.
- [Probability, Patterns & Decision Making](https://broomva.tech/writing/probability-patterns-and-decision-making): Expected value as a framework for reasoning through uncertain decisions — from marble games to information asymmetry.
- [Control Systems as Self-Engineering](https://broomva.tech/writing/control-systems-as-self-engineering): The same equations that stabilize machines describe how we stabilize ourselves — feedback loops, phase margins, and disturbance rejection as a language for human self-optimization.

## Notes

- [Autonomy is governance](https://broomva.tech/notes/autonomy-is-governance): The degree of autonomy you can grant an agent is directly proportional to the quality of your governance infrastructure.
- [Harness over prompting](https://broomva.tech/notes/harness-over-prompting): Prompt quality matters, but harness quality is what keeps agent outputs reliable once stakes increase.
- [A release rhythm that works](https://broomva.tech/notes/release-rhythm): One weekly long-form post plus short daily notes creates enough surface area for learning, distribution, and iteration.
- [Agent UI needs state visibility](https://broomva.tech/notes/ui-needs-state-visibility): Operators trust agents when they can inspect current state and next action without digging through logs.

## Prompts

- [AI-Native Platform Architect](https://broomva.tech/prompts/ai-native-platform-architect): Greenfield AI-native platform design from first principles — primitives (memory, sessions, agents, artifacts, runtime), ontology, runtime architecture, UX, onboarding, and implementation sequence. Refuses to reduce to a SaaS dashboard with AI added on top.
- [Bstack-Native AI Platform Architect](https://broomva.tech/prompts/bstack-native-ai-platform-architect): AI-native platform architecture from first principles, with bstack P1-P20 as the substrate, filesystem+bash/skills+markdown+hooks+bridge as the product shape, and aiOS as the build engine. Combines deep ontology design with executional rigor (knowledge-graph memory, parallel agents, dep-chain analysis, cross-model review) for greenfield platforms and new ecosystem repos.
- [Ecosystem Repo Architect](https://broomva.tech/prompts/ecosystem-repo-architect): Deep architectural reasoning workflow for creating new repositories in the Broomva ecosystem using knowledge-graph-memory, agentic control layer, parallel agents, and step-by-step dependency chain analysis.
- [Symphony vs Direct Autonomous Execution: Operational Value Assessment](https://broomva.tech/prompts/symphony-vs-direct-autonomous-execution-operational-value-assessment): Deep comparative analysis for evaluating /symphony orchestration vs direct autonomous agent execution, grounded in actual codebase behavior and measurable outcomes.
- [Autonomous Senior Engineer](https://broomva.tech/prompts/autonomous-senior-engineer): Autonomous implementation mode with full dependency-chain reasoning, parallel agents, and CI/CD validation
- [Broomva Ecosystem Coherence Audit](https://broomva.tech/prompts/broomva-ecosystem-coherence-audit): End-to-end audit of broomva.tech, Broomva CLI, and /broomva ecosystem to close the reasoning/execution loop
- [bstack Control Harness Bootstrap](https://broomva.tech/prompts/bstack-control-harness-bootstrap): Implement and validate the bstack control harness — self-improving governance layer with hooks, pre-commit gates, consciousness stack, and knowledge graph indexing
- [Agent-Native Architecture Analysis](https://broomva.tech/prompts/agent-native-architecture): Deep architectural analysis prompt for designing apps where frontend state and agent state are unified. The agent is the app; the chat is just one interface.
- [Agent Observability & Evaluation](https://broomva.tech/prompts/agent-observability-evaluator): Instrument agent behavior with tracing, connect to evaluation frameworks, and bridge traces to the knowledge graph for persistent observability.
- [Agentic Development Loop](https://broomva.tech/prompts/agentic-development-loop): Full autonomous development workflow: branch, implement, test, document, commit, PR, and verify. Integrates control metalayer, knowledge graph, and CI checks at every step.
- [Autonomous Orchestration Loop](https://broomva.tech/prompts/autonomous-orchestration-loop): Full autonomous feature delivery: activate control metalayer, knowledge graph, and consciousness stack, then use Symphony with worktrees and parallel agents to complete Linear tickets, verify PR checks, and merge — all hands-off.
- [Brutally Honest Advisor](https://broomva.tech/prompts/brutally-honest-advisor): Unfiltered strategic advisor persona that challenges assumptions, exposes blind spots, and delivers direct truth over comfort. No flattery, no softening.
- [Code Review Agent](https://broomva.tech/prompts/code-review-agent): Structured code review prompt with security, performance, and style gates. Produces actionable findings with severity levels.
- [Codebase Deep Analysis](https://broomva.tech/prompts/codebase-deep-analysis): Comprehensive codebase traversal and analysis prompt. Reads full files, traces dependency chains, evaluates architecture patterns, and produces actionable reports.
- [Compact: Metalayer Bootstrap with Kernel](https://broomva.tech/prompts/compact-metalayer-bootstrap): Compact command to bootstrap and self-update CLAUDE.md, AGENTS.md, and .control/ wired to the agentic control kernel.
- [Consciousness Stream Synthesis](https://broomva.tech/prompts/consciousness-stream-synthesis): Consolidate the control metalayer, Obsidian knowledge graph, and conversation logs into a self-evolving consciousness architecture document. Meta-cognition for agent systems.
- [Content Creation Pipeline](https://broomva.tech/prompts/content-creation-pipeline): End-to-end content creation workflow from idea to published blog post with structured research, drafting, and editing phases.
- [Control Metalayer Initialization](https://broomva.tech/prompts/control-metalayer-init): System prompt for bootstrapping a control-system metalayer in any repository, establishing governance primitives for autonomous agent development.
- [Deep Research Agent](https://broomva.tech/prompts/deep-research-agent): Multi-source research prompt with citation tracking, claim verification, and structured synthesis for comprehensive analysis.
- [Deep Thinking Directive](https://broomva.tech/prompts/deep-thinking-directive): Meta-directive for rigorous engineering. Forces step-by-step dependency chain analysis, best-practice research, and premium-quality implementation across any task.
- [Knowledge Graph Documentation](https://broomva.tech/prompts/knowledge-graph-documentation): Structure a repository's documentation as a traversable Obsidian knowledge graph with wikilinks, MOC navigation, tags, and frontmatter. The repo itself becomes the vault.
- [Metalayer Bootstrap with Control Kernel](https://broomva.tech/prompts/metalayer-bootstrap-with-kernel): Bootstrap and self-evolve a repository's metalayer (CLAUDE.md, AGENTS.md, .control/) wired to the agentic control kernel for autonomous governance with drift detection and safe self-update.
- [Ontology & State Design](https://broomva.tech/prompts/ontology-state-design): Design ontology definitions and global app state schema for a business domain. Produces typed state that unifies UI rendering and agent reasoning.
- [Platform Revamp Planner](https://broomva.tech/prompts/platform-revamp-planner): Plan a full platform revamp: decompose into turborepo apps, migrate existing code, create dependency-ordered tickets, and orchestrate parallel agent execution.
- [PR Review & Resolution Agent](https://broomva.tech/prompts/pr-review-agent): Systematic PR comment resolution workflow. Reads review comments, triages by type, fixes or rejects with justification, and ensures all CI checks pass before merge.
- [Prompt Engineering Guide](https://broomva.tech/prompts/prompt-engineering-guide): Meta-prompt for creating effective, structured prompts. Use this to generate new prompts that follow best practices for clarity, specificity, and reproducibility.
- [Rust Agent Architect](https://broomva.tech/prompts/rust-agent-architect): Design a Rust-native agent runtime with daemon architecture, provider abstractions, and harness-driven control loops. Think Claude Code meets systems programming.
- [Secure Refactor Agent](https://broomva.tech/prompts/secure-refactor-agent): Refactor brittle data pipelines into properly typed, secure, cached, and stateful abstractions. Traces the full dependency chain before changing anything.
- [Security Attack Surface Assessment](https://broomva.tech/prompts/security-attack-surface): Systematic attack surface analysis targeting API core, state layer, and middleware. Produces a protection wrapper with fixes prioritized by exploitability.
- [Technical Valuation Report](https://broomva.tech/prompts/technical-valuation-report): Generate a comprehensive technical valuation report for a codebase. Estimates construction cost using token count, engineering time, and market rate proxies.
- [UX Interaction Design](https://broomva.tech/prompts/ux-interaction-design): Design micro-interactions, transitions, and motion patterns for AI-native interfaces. Collapsible docks, liquid glass modals, state-preserving navigation.

## APIs

### Prompts API
- `GET https://broomva.tech/api/prompts` — List all prompts. Query params: `category`, `tag`, `search`.
- `GET https://broomva.tech/api/prompts/[slug]` — Get a single prompt by slug. Returns title, content, category, tags.

### Chat API
- `POST https://broomva.tech/api/chat` — AI chat endpoint. Requires authentication. Supports streaming responses via Server-Sent Events.

### Content API
- `GET https://broomva.tech/sitemap.xml` — Full sitemap with all pages and content
- `GET https://broomva.tech/llms.txt` — Concise site overview for LLMs
- `GET https://broomva.tech/robots.txt` — Crawler directives
- `GET https://broomva.tech/<kind>/<slug>.md` — Raw markdown for any content page (kind: writing, projects, notes, prompts)

## Features

- **AI Chat**: authenticated streaming chat using the deployed model catalog; tools and provider routes depend on active configuration and credentials
- **Knowledge Graph**: Obsidian vault integration for connected note-taking
- **Memory**: persistence is deployment-dependent and not represented as universally available
- **Content Platform**: MDX-based writing, projects, notes with full-text search
- **Prompt Library**: Versioned, categorized prompts with API access

## Contact

- Website: https://broomva.tech
- Email: contact@broomva.tech
- Operator: Carlos D. Escobar-Valbuena, operating as Broomva