Research Notes
-
March 18, 2026CAMBRIAN: What If the Spec Is the Organism?What if instead of patching code, we evolved the specification and regenerated the agent from scratch? The spec is the genome, the code is the phenotype, and economic viability is the fitness function. An agent that can't pay its own bills — compute, LLM calls, storage, bandwidth — goes extinct. Prediction markets as the first income strategy. CAMBRIAN-SPEC-001: the 300-line generative specification for a self-sustaining software lineage.
-
March 17, 2026It Rewrote Itself: Loom's First Autonomous Self-ModificationGen-72: a ClojureScript agent autonomously modified its own source code, verified the change, and promoted it to master in 56 seconds. The real story is the 16 failures that got us there — six infrastructure bugs, three models tested, and the lessons each one taught.
-
March 16, 2026First Light: Loom's Self-Modification Pipeline in 2,214 LinesThe MVP field report: 2,214 lines of ClojureScript, 90 tests, 17 generations run. From early failures to a 5/5 stable pipeline. What we built, what broke, what's still missing, and the road to recursive self-improvement.
-
March 13, 2026The program.md Protocol: Steering Self-Improvement with a ContractHow do you steer a self-improving agent? With a collaboratively written contract. The refined execution model: user and Prime draft a
program.md, an autonomous Lab executes it, Prime verifies independently, and git tracks every generation. -
March 12, 2026The Prime and the Lab: Recursive Self-Improvement for Coding AgentsA ClojureScript agent running in a container, modifying itself, and proving the modification in a second container before promoting it. Both the argument and the spec. No frameworks, no build step — just a loop, a contract, and two containers.
-
March 11, 2026Loom in Lean: Bootstrapping a Verified Self-Improving AgentWe argued for Elixir, then found its limits. Lean's type checker is an incorruptible judge for self-improvement. Memory is the beachhead. Here's what's hard about it, and what kills it.
-
March 11, 2026The Autoresearch Pattern: A Blueprint for Self-Improving AgentsKarpathy's autoresearch reveals the structure that makes self-improvement work: three separations, a keep/revert loop, and well-chosen fixed points. The blueprint for agents that get better overnight.
-
March 5, 2026Loom: The Case for Building on OpalAn existing Elixir port of the Pi coding agent already solves the hardest architectural problem. One architect, frontier models, five weeks. Here's the case.
-
March 5, 2026Porting OpenClaw's Core to ElixirA practical assessment: port the core gateway and agent loop to Elixir, keep the TypeScript channels, and bridge them. 3–4 months for humans, 4–6 weeks with coding agents.
-
March 5, 2026Runtime Self-Modification: Beyond TypeScriptWhy building a self-modifying agent in TypeScript is writing letters to your future self. Common Lisp offers depth, Elixir/BEAM offers safety — and the hybrid might be the real answer.
-
March 5, 2026Making OpenClaw Self-Aware: How Can an AI Agent Patch Its Own Code?Five approaches to giving an AI agent a coherent mental model of its own codebase — from source-aware skills to a dedicated dev agent — and the security tradeoffs of each.
-
February 13, 2026What if the LLM Thought in S-Expressions?A thought experiment: designing an AST-native code editing protocol for LLMs using Lisp's homoiconicity. Inspired by Can Bölük's hashline work.
-
February 13, 2026Lunar Habitat Life Support SystemsA comprehensive research report mapping the companies, technologies, and supply chains for atmospheric control and CO₂ scrubbing in next-generation lunar stations.