evo-hq/evo

Discover initializes an evo workspace for the current repository by exploring the codebase, proposing optimization dimensions, constructing a benchmark, and running a baseline experiment. Use it when starting a new evo run or instrumenting a project for autonomous optimization.

Evo Report provides read-only reports from recorded evo workspace state, including score charts, frontier candidates, and status summaries. Use it when someone asks what happened overnight, what improved, or wants a quick chart without opening the dashboard.

Evo Subagent Protocol defines the brief shape and execution loop that evo optimization subagents follow when dispatched from /optimize. Orchestrators use it to write valid briefs and debug subagent behavior; subagents use it to drive required verifier checks and read-edit-run iteration.

Finetuning selects or diagnoses a training move—SFT, LoRA, DPO/KTO/ORPO, RFT, GRPO/PPO/RLOO, RLHF—based on reward shape and literature, then gates runs with smoke tests, early stopping, and diagnostics. Use it when fine-tuning, post-training, reward design, or weight updates come up.

Infra Setup guides switching the evo backend between local worktrees, pool slots, and remote providers like Modal, E2B, or SSH. It checks prerequisites, verifies auth, and shows the exact config commands so experiments run on the intended infrastructure.

Optimize drives structured autoresearch iteration after evo:discover and the baseline commit, orchestrating subagents to run experiments and improve the current best frontier. Use it when the user invokes /evo:optimize or asks to try ideas, variants, or continue an evo search.

Optimize runs the evo optimization loop with parallel subagents until interrupted or the stall limit is reached. Use it after a baseline experiment is committed to evolve code changes against benchmarks and prune dead branches.

Ship turns a finished evo run into a clean, mergeable change by distilling the best-scoring experiment down to its minimal diff, restoring test integrity, and opening a pull request or merging locally. Use when asked to land, merge, ship, or turn a finished optimization into a reviewable change.