Slash Commands

Slash commands are available inside the Hypatia REPL. They map to research workflows, project utilities, and commands from the bundled Pi packages. Type /help inside the REPL for Hypatia’s grouped command list.

Research workflows

Command	Description
/deepresearch <topic>	Run a thorough, source-heavy investigation and produce a research brief with inline citations
/lit <topic-or-lab-or-author>	Run a literature review on a topic, lab, PI, or author
/review <artifact>	Run an internal research critique with likely objections, severity, and a concrete revision plan
/audit <item>	Compare a paper’s claims against its public codebase for mismatches, omissions, and reproducibility risks
/replicate <paper>	Plan a replication workflow for a paper, claim, or benchmark; execute only after an explicit environment choice
/recipe <task-or-paper>	Find ranked, implementable ML training recipes backed by papers, datasets, docs, and code
/compare <topic>	Compare multiple sources and produce a matrix of agreements, disagreements, and confidence
/draft <topic>	Turn research findings into a paper-style draft
/autoresearch <idea>	Run persistent benchmark experiments with approved contracts, observed receipts and pause/resume
/summarize <source>	Summarize a paper, report, README, local artifact, or PDF with the source kept on disk instead of in context

Workflow prompts can call the bundled agents (evidence-scout, scribe, auditor, critic, reviser) through Pi’s subagent tool when delegation helps; narrow tasks stay lead-owned.

Project and session

Command	Description
/log	Write a durable session log with completed work, findings, open questions, and next steps
/help	Show grouped Hypatia commands and prefill the editor with a selected command
/init	Bootstrap AGENTS.md and session-log folders for a research project
/idea <research goal>	Generate bounded research ideas in chat with native Idea Forge
/experiments	Native idea creation, experiment setup, review gates, receipts, pause/resume, trees and provenance
/outputs	View grouped reports/provenance in the TUI, search text, or copy artifact paths
/sources	Browse citations in the latest assistant answer; inspect identifiers and explicitly stated verification status, then open or copy a URL
/compare-provenance	Compare recorded source coverage and verification metadata from two completed research runs
/model-routes	Configure ordered subagent models and choose Pi-override or parent-model inheritance
/tools	Browse public research tools with their source and parameter summary
/service-tier	View or set the provider service tier override for supported models

Reading research outputs

/outputs groups files under papers/, outputs/, experiments/, and notes/ into Reports, Provenance, Drafts, and Other artifacts. Reports are offered first, and each group is sorted newest first. Group labels describe file locations/types; they do not certify completion or verification.

Selecting Markdown opens a local reader; other supported text formats open as plain text. Use arrows, Page Up/Page Down, Home/End, or the fullscreen mouse wheel to scroll. / or Ctrl+F opens a literal, case-insensitive search of rendered text; Enter applies it, and n/N moves between matching rows. Escape cancels search editing, then returns from the reader to the list.

Press p to switch to a report/provenance sidecar with the same stem in the same directory, when present. Press c to copy the current file’s absolute path through Pi’s clipboard support. Local viewing/copying does not submit a model request or replace composer text. Read in chat, or r in the reader, explicitly prefills the file path; you review and submit the prompt yourself.

Previews are limited to 2 MiB. PDFs and files over that limit offer Copy path and Read in chat. Artifact discovery skips symlinks; previews stay inside the workspace’s research folders. Clipboard availability follows the terminal/platform’s support.

Package commands

Command	Package	Description
/btw <question>	pi-btw	Ask a side question while the main research agent is busy
/search	pi-web-access	Browse stored web search results from the current session
/websearch [queries]	pi-web-access	Open the search curator to run and review searches yourself
/curator [on|off]	pi-web-access	Toggle the search curator workflow
/subagents	pi-subagents	Inspect configured agents
/run <agent> [task] [--bg] [--fork]	pi-subagents	Run one agent
/hotkeys, /new, /quit	Pi	Show keyboard shortcuts, start a new session, quit

Research delegation

The current runtime does not provide /chain or /parallel commands. Research workflows call subagent directly for one child, or write a workflow script for parallel or sequential work to outputs/.plans/<slug>-workflow.js and run it by calling subagent with that path. A file works with every model; an inline ```js workflow block with workflow: true fails when a model puts the block in its hidden reasoning instead of the reply.

return await runs.all([
  { key: "papers", agent: "evidence-scout", task: "Read outputs/.plans/<slug>-papers.md.", output: "outputs/.drafts/<slug>-research-papers.md" },
  { key: "web", agent: "evidence-scout", task: "Read outputs/.plans/<slug>-web.md.", output: "outputs/.drafts/<slug>-research-web.md" },
]);

{ "workflow": "./outputs/.plans/<slug>-workflow.js", "async": true, "globalConcurrencyLimit": 4 }

For sequential work, await runs.run(key, {agent, task, output}) before starting the next step. Legacy top-level tasks, chain, and parallel inputs are rejected. runs.all returns an ordered array, including ordinary child failures; inspect each result’s ok and returned output references.

An async launch receipt is not completion: consume the result and verify child output paths before synthesis or dependent review. Ordinary background subagents notify their parent; bg_wait is for background work without a native completion notification.

Running workflows from the CLI

Every research workflow can also be run from the command line, for example hypatia deepresearch "topic". See CLI Commands.

For experiment setup in chat use /experiments create <goal>; use create-json <path> for saved contracts. import <idea-file> prepares a contract in chat; import-json <idea-file> --contract <path> loads one directly. review <slug> prefills a review-only chat request. New contracts require a critic before benchmarks and an auditor before closure.

/idea <goal> starts native Idea Forge preparation in chat; /experiments generate <goal> is an alias with Hypatia auth and no clone; ideas browses generations/proposals. Use forge-status, forge-approve, forge-pause, forge-resume, or forge-recover with a generation slug. Model API generation and benchmark execution have separate approved contracts.

Tree controls are /experiments tree create, list, status <slug>, approve <tree> <node>, fork <tree> <parent> <child> <hypothesis>, promote <tree> <node>, pause, resume, recover, close-node <tree> <node>, close, and export. Tree creation requires a clean Git repository root; each node runs in a detached worktree and shares the tree’s trial/time/parallel-run budgets. /autoresearch tree <tree> <node> continues a selected node. Promotion records a reviewed winner without merging or changing the checkout.