CLI Commands — Feynman Docs

This page covers the dedicated Feynman CLI commands and flags. Workflow commands like feynman deepresearch are also documented in the Slash Commands reference since they map directly to REPL slash commands.

Core commands

Command	Description
`feynman`	Launch the interactive REPL
`feynman chat [prompt]`	Start chat explicitly, optionally with an initial prompt
`feynman help`	Show CLI help
`feynman setup`	Run the guided setup wizard
`feynman setup preview`	Install or verify preview dependencies
`feynman doctor`	Diagnose config, auth, Pi runtime, and preview dependencies
`feynman status`	Show the current setup summary (model, auth, packages)
`feynman rank "topic"`	Rank papers for deciding what to read first, with transparent citation, method, reproducibility, and provenance evidence
`feynman paper <id-or-title>`	Resolve legal full-text access candidates for one paper and optionally fetch source-specific text

Paper access commands

Command	Description
`feynman paper 10.7717/peerj.4375`	Resolve OpenAlex, DOI, publisher/repository, and Europe PMC access candidates
`feynman paper pmid:29456894`	Resolve a PubMed/PMC-indexed paper through exact OpenAlex PMID/PMCID metadata and Europe PMC candidates
`feynman paper 2309.08600 --fetch-full-text`	Fetch text through source-specific APIs when available and write bounded access artifacts
`feynman paper "paper title" --json`	Search OpenAlex by title and print a machine-readable access summary

Paper access writes <slug>-paper-access.md and <slug>-paper-access.json. It records access candidates from OpenAlex, DOI, PMID/PMCID, arXiv/alphaXiv, and Europe PMC. It does not bypass paywalls and does not write raw full-text bodies to artifacts.

PaperRank commands

Command	Description
`feynman rank "topic"`	Fetch OpenAlex works and rank papers for a topic
`feynman rank "topic" --limit 20`	Limit the candidate paper count
`feynman rank "topic" --expand-citations 2`	Add cited and citing works to the local graph before scoring graph prestige
`feynman rank "topic" --full-text-top 3`	Fetch source-specific full text for top candidates, add section-aware rubric evidence, and rescore
`feynman rank "topic" --critique-top 5`	Write research-critique strengths, concerns, and follow-up questions for top ranked papers
`feynman rank "topic" --preference-file preferences.json`	Evaluate rank agreement against researcher read-order preferences
`feynman rank "topic" --reproduction-notes reproduction-notes.json`	Record completed reproduction outcomes separately from planned replication checks
`feynman rank "topic" --synthesis-top 7`	Choose how many ranked papers enter the bounded model-synthesis packet
`feynman rank "topic" --synthesize`	Ask the recommended available non-Pro research model to write `<slug>-model-synthesis.md` and print the selected model
`feynman rank "topic" --synthesize --model provider/model`	Run model synthesis with an explicit non-Pro model for this command
`feynman rank "topic" --synthesize --synthesis-model provider/model`	Run model synthesis with an explicit non-Pro model without changing the chat model flag
`feynman rank "topic" --output-dir outputs`	Choose where artifacts are written
`feynman rank "topic" --json`	Print a compact JSON summary after writing artifacts

PaperRank writes a ranked brief, normalized paper/score JSONL, a score audit, citation/field context, graph explorer, rank-sensitivity data, and provenance by default. Optional flags add research critique, empirical preference calibration, completed reproduction notes, source-specific full-text enrichment, citation-neighborhood expansion, or model synthesis. The CLI output, JSON summary, generated synthesis, and provenance record the selected model and whether it came from the recommendation path or an explicit override. The score separates topical relevance, citation impact, local graph prestige, citation velocity, methodology screening, and reproducibility screening. The score audit explains per-paper weights, contribution math, visible evidence, missing components, rubric gaps, field roles, and critique status. Preference files and reproduction notes are treated as external evidence; without them, PaperRank labels those checks as not provided without writing extra calibration or reproduction files. The synthesis packet and prompt are bounded model inputs that omit raw full text and are written only when synthesis is requested. The graph explorer is an inspection view and does not embed raw full-text bodies.

Model management

Command	Description
`feynman model list`	List available models in Pi auth storage
`feynman model login [id]`	Authenticate a model provider with OAuth or API-key setup
`feynman model logout [id]`	Clear stored auth for a model provider
`feynman model set <provider/model>`	Set the default non-Pro model for all sessions

These commands manage your model provider configuration. The model set command updates ~/.feynman/settings.json with the new default. It accepts either provider/model-name or provider:model-name; run feynman model list first and choose a non-Pro model ID from that output. Running feynman model login google or feynman model login amazon-bedrock routes directly into the relevant API-key setup flow instead of requiring the interactive picker.

AlphaXiv commands

Command	Description
`feynman alpha login`	Sign in to alphaXiv
`feynman alpha logout`	Clear alphaXiv auth
`feynman alpha status`	Check alphaXiv auth status
`feynman alpha search "query"`	Search papers through Feynman’s bundled alphaXiv client
`feynman alpha get <id-or-url>`	Fetch paper content and local annotations
`feynman alpha ask <id-or-url> "question"`	Ask a question about a paper
`feynman alpha code <github-url> [path]`	Inspect a paper repository
`feynman alpha annotate ...`	Read, write, list, or clear local paper notes

AlphaXiv authentication enables Feynman to search and retrieve papers, access discussion threads, and pull citation metadata. Use feynman alpha ... for shell access so Feynman runs its bundled patched alphaXiv client.

Package management

Command	Description
`feynman packages list`	List supported optional research packages and their install status
`feynman packages install <preset>`	Install an optional package preset
`feynman update [package]`	Update installed packages, or a specific package by name

Use feynman packages list to see which optional research-continuity packages are available on your platform and which are already installed. The default install keeps only the research essentials in core. Install optional presets one by one when they directly support an active research workflow.

Utility commands

Command	Description
`feynman search status`	Show Pi web-access status and config path

REPL hotkeys

Inside the interactive REPL, use /hotkeys to show the live keyboard map. The default reasoning controls are:

Hotkey	Action
`Shift+Tab`	Cycle thinking/reasoning level
`Ctrl+T`	Toggle thinking block visibility

Workflow commands

All research workflow slash commands can also be invoked directly from the CLI:

feynman deepresearch "topic"
feynman lit "topic-or-lab"
feynman review artifact.md
feynman audit 2401.12345
feynman replicate "claim"
feynman recipe "fine-tune a small model for math reasoning"
feynman compare "topic"
feynman draft "topic"

These are equivalent to launching the REPL and typing the corresponding slash command.

Flags

Flag	Description
`--prompt "<text>"`	Run one prompt and exit (one-shot mode)
`—model <provider/model	provider:model>`
`--thinking <level>`	Set thinking level: `off`, `minimal`, `low`, `medium`, `high`, `xhigh`
`--cwd <path>`	Set the working directory for all file operations
`--session-dir <path>`	Set the session storage directory
`--new-session`	Start a new persisted session
`--alpha-login`	Sign in to alphaXiv and exit
`--alpha-logout`	Clear alphaXiv auth and exit
`--alpha-status`	Show alphaXiv auth status and exit
`--doctor`	Alias for `feynman doctor`
`--setup-preview`	Alias for `feynman setup preview`