Search for arXiv papers by text, title, author, or category.

Returns paginated results with paper metadata and relevance scores. Use cursor from previous results to get the next page.

Optionally provide profile_slug to get profile-ranked results with ranking explanations on each result.

Response shape: {"results": {"items": [...], "page_info": {...}}, "ranker_snapshot": ...}

Without profile_slug: items include triage_state and collection_slugs; ranking_explanation is null; ranker_snapshot is null.

With profile_slug: items additionally include ranking_explanation and the response includes a ranker_snapshot capturing ranker config.

Browse recently announced arXiv papers, optionally filtered by category.

Use time_basis to select ordering: announced, submitted, or updated. Note: Many papers may not have announced_date populated. If results are empty with the default time_basis='announced', try time_basis='submitted'.

Optionally provide profile_slug to get profile-ranked results with ranking explanations on each result.

Response shape: {"results": {"items": [...], "page_info": {...}}, "ranker_snapshot": ...}

Without profile_slug: items include triage_state and collection_slugs; ranking_explanation is null; ranker_snapshot is null.

With profile_slug: items additionally include ranking_explanation and the response includes a ranker_snapshot capturing ranker config.

Find papers related to one or more seed papers via lexical similarity.

Accepts a single arXiv ID or a list of IDs. When multiple seeds are given, results are merged and deduplicated, keeping the highest score for each paper.

Get metadata for a single paper by its arXiv ID.

Set the triage state for a paper.

Valid states: seen, shortlisted, dismissed, read, cite-later, archived, unseen.

Add a paper to a collection. Creates the collection if it doesn't exist.

Create a monitored search that tracks new papers matching your query.

Read the watch://{slug}/deltas resource to see papers added since your last check.

Add a signal to an interest profile.

Signal types: seed_paper (arXiv ID), saved_query (query slug), followed_author (author name), negative_example (arXiv ID).

Add multiple signals to an interest profile in one call.

Each signal dict must have: signal_type, signal_value. Optional per-signal: reason.

Returns a summary with counts and per-signal results. Continues on individual signal errors (partial success is OK).

Create a new interest profile.

Returns the profile summary with slug, name, signal_count, and timestamps.

Generate signal suggestions for an interest profile.

Examines workflow activity (triaged papers, frequent queries, recurring authors) to suggest new signals. With auto_add=True, adds suggestions as pending signals automatically.

Returns candidates list and added_count.

Trigger OpenAlex enrichment for a paper to get topics, citations, and related works.

Set refresh=True to re-enrich even if data exists within the cooldown window.

Get paper content at the requested fidelity level.

Retrieves paper content as abstract, HTML, or PDF-derived markdown. Use variant='best' (default) to get the highest-quality available format, which tries HTML first then falls back to PDF markdown.

Valid variants: 'abstract', 'html', 'pdf_markdown', 'best'

Returns content with provenance metadata (source, backend, license). For non-abstract variants, respects per-paper license restrictions.