Configure the project's skill stack and supervision preferences. Reads the curated registry in `skillpacks/skill_dictionary.yaml`, asks a short preset-first set of questions about workflow, dependency tolerance, autonomy style, and resource policy, then writes or updates `.co-researcher/skills.yaml`. Trigger phrases: "customize my stack", "configure skillpacks", "set up my skills", "choose presets", "configure supervision and packs", "personalize this project".
Three modes. Session mode (default): extracts generalizable lessons from RESEARCH.md and git history at session end; lessons that imply a new or significantly changed skill are handed off to skill-creator. Personalize mode: searches the skills registry via `npx skills find`, reads the target skill(s), checks compatibility and scope overlap against installed skills, interviews the user to understand what they want and what to skip, then creates or improves skills using skill-creator. Registry mode: curates `skillpacks/skill_dictionary.yaml` and `skillpacks/presets/*.yaml` by assessing external packs, judging necessity/compatibility, and recommending subsets. Create mode: designs a brand- new skill from scratch using skill-creator. Never edits SKILL.md directly — all changes go through skill-creator's draft→test→iterate loop, human merges. Trigger phrases: "end session", "extract lessons", "personalize my skills", "integrate this skill", "update skillpack", "find a skill for", "create a skill", "improve skill",
Runs ML experiments reproducibly — single runs or autonomous BFS batches. Single mode: isolated venv, time-budgeted, failure-handled, logs to RESEARCH.md. BFS mode (opt-in): designs N hypotheses, runs each for a fixed budget, compares via a single verifiable metric, keeps improvements and git-resets failures — fully autonomous until done. Respects the RESEARCH.md supervision policy for notifications, approvals, and stop limits. Trigger phrases: "run experiment", "train model", "explore design space", "find best config", "autoresearch".
Autonomous research agent that reads RESEARCH.md, infers what's needed, dynamically adjusts TODOs, and delegates to the right skill. Supports opt-in BFS mode for autonomous design space search. Respects a configurable supervision policy (presets: manual / checkpointed / autonomous / wild) governing notifications, approval gates, resource limits, and idea-change handling. Proactively surfaces gaps and asks before acting. Trigger phrases: "start research", "continue project", "what's next?", "explore design space", "autoresearch".
Adversarial critique of a research draft or paper using an isolated critic context. The critic must have no access to the current conversation history — context isolation, not model identity, is what makes the review independent. Use when a draft is complete or partially complete, after each writing round, or when `co-write` invokes it automatically. Returns FATAL/MAJOR/MINOR issues, a structured score (0–10), and a PROCEED/REFINE/PIVOT recommendation. Caps at 4 rounds. Trigger phrases: "review paper", "critique draft", "check citations", "adversarial review".
Configure the project's supervision policy in RESEARCH.md. Uses a preset-first flow (`manual`, `checkpointed`, `autonomous`, `wild`), then lets the user adjust notification events, approval gates, stop limits, resource rules, and idea-change handling. Trigger phrases: "configure supervision", "set supervision", "automation settings", "change autonomy", "/supervision".
Drafts or revises a publication-ready research paper from RESEARCH.md, producing paper.md and refs.bib. Builds a clear claim-evidence narrative, writes in professional academic style, marks ungrounded claims [UNGROUNDED] and unverified citations [UNVERIFIED] inline, then auto-invokes co-review. Trigger phrases: "write paper", "draft paper", "revise paper", "start writing".