Research
13 records filed under this subject, newest first, each with what it is, what it cost and where the builder got stuck.
Records
13
131AntOmniEvoA Python framework that runs an evolution loop over a directory of your files: a coding agent rewrites them from the failure trajectories of the previous candidate, parent and child are scored on the same training batch, and only an improvement that clears a threshold is re-scored on the validation set and kept.ClaudeClaude Code
123agent-memoryA long-term memory runtime for AI agents that keeps plain Markdown files as the single source of truth, ranks them locally without calling a model, answers recall with file paths the agent opens one level at a time, writes at conversation boundaries rather than on the agent’s initiative, and runs an independent sleep-time layer that may add and update on its own but can only ever file a deletion as a proposal — one store shared by Claude Code, Codex CLI and Hermes, with no API key.ClaudeClaude CodeCodex
117sepiaA portable de-AI writing skill: four operations over one canonical rules file, narrative architecture repaired before word choice on fiction, a thin rule file matched to the venue on professional prose, and every rule labelled as measured, consulted or the project’s own inference.ClaudeClaude CodeCodex+4
120EaselAn open-source content workbench for social media creators: one agent runs the whole loop — aggregate the hot lists, plan a topic, generate the copy, the cards, the voice and the video, publish the finished file to an account that is already logged in on seven Chinese platforms, then read the numbers back into the account profile that shaped the next round.ClaudeClaude CodeGemini+3
116LemmalogA Rust Datalog engine that treats agent memory as a deductive database rather than a bigger vector store: facts asserted at the extraction boundary, stratified rules deriving closures and temporal views, provenance back to the source episode on every derived fact, and views maintained one epoch at a time — served to Claude Code and Kimi CLI as twelve MCP tools.ClaudeClaude CodeGPT
075The Fable MethodHow Claude Fable 5 worked, written down as four skills any model can run, plus the adversarial eval that keeps them honest: fourteen trap fixtures, blind judges that diff and execute rather than read reports, and fifteen rounds of results with the failures and the nulls left in.ClaudeClaude Code
073EngramA learning engine that installs into a coding agent: a curriculum architect breaks a topic into a first-principles concept map, a tutor makes you predict, attempt and explain before it explains, a blind assessor grades your verbatim free recall and writes a receipt for every verdict, and a deterministic FSRS-4.5 core in one Python file decides when each concept comes back — with explorable HTML built only for the concepts whose content rewards manipulation.ClaudeClaude CodeCodex+2
063book-to-skillA converter that reads a document — one file, a folder, a glob or a list of paths, in PDF, EPUB, DOCX, HTML, RTF, MOBI or plain text — and writes an agent skill: a core file of mental models with a chapter index, one file per chapter that loads only when a question touches it, and a glossary, a patterns file and a cheatsheet beside them.ClaudeClaude CodeCursor+2
055LeanCTXA local layer that sits beside a coding agent and decides what reaches the model: file reads are compressed and cached, command output is compressed by per-command rules, session findings persist across chats, and a local proxy rewrites each request without breaking the provider’s prompt cache — with a savings ledger, a budget and a dashboard for what it measured.ClaudeClaude CodeCursor+2
047autoresearchGive an AI agent an LLM training script, a fixed five-minute budget and no supervision — then see what it found while you slept.Claude CodeCodex
033MiroFishA prediction engine that builds a parallel world of hundreds of AI agents out of a document you upload, then runs it forward to see what happens next.ClaudeGemini
017BettaFishA public-opinion analysis system in which several kinds of research agent are made to argue with each other on purpose, so the output is not one model’s opinion written up at length.
015SWE-agentA research project that argued the model was never the bottleneck — the interface was.GPTClaude