# Lacuna Research MCP ### 🔬 Ground your coding agent in novel ideas, papers, and the ML landscape Lacuna Research MCP gives AI researchers' coding agents: - **Novel research proposals.** Explore novel research ideas generated with [Alien Science](https://openreview.net/pdf?id=XZWkDET1ia). - **Research directions.** Navigate concept clusters with linked papers, authors, and proposals. - **Agent-ready literature.** Search recent papers in markdown with source links. - **Researcher intelligence.** Trace authors, publications, directions, impact, and related researchers. - **Landscape mapping.** Compare venues, institutions, leading researchers, and publication activity. [Lacuna](https://lacuna.tiptreesystems.com), built by [Tiptree Systems](https://tiptreesystems.com), is a research map of machine learning: a heterogeneous knowledge graph linking papers, research directions, authors, venues, institutions, and generated research proposals, with a source trail from every derived object back to the exact paper and page that produced it. Its pipeline reconciles scholarly records from OpenAlex, OpenReview, DBLP, and arXiv; extracts concept elements from paper text and clusters them into research directions ([Lacuna paper](https://arxiv.org/abs/2606.26246)); and samples novel research proposals from those directions with [Alien Science](https://arxiv.org/abs/2603.01092). The map spans more than **730,000 papers**, **190,000+ author profiles**, **38,000+ research directions**, and **3,000+ research proposals** built from over **15 million concept elements** — and it grows continuously as new arXiv and other AI papers are ingested. [Install](#install) · [First use](#first-use) · [Tools](#what-it-exposes) · [API reference](#wrapped-apis) · [Configuration](#environment-variables) ## Install The easiest way to install Lacuna Research MCP is to ask your coding agent, such as Codex or Claude Code: > Install and configure the `lacuna-research-mcp` package from PyPI for this client. For manual setup, the instructions below use [`uvx`](https://docs.astral.sh/uv/guides/tools/) to run the latest tagged release from PyPI. Install [uv](https://docs.astral.sh/uv/getting-started/installation/) first; Lacuna Research MCP requires Python 3.11 or newer. ### Codex Add the server with the Codex CLI: ```bash codex mcp add lacuna-research -- uvx lacuna-research-mcp ``` Alternatively, add the following to `~/.codex/config.toml` (or to `.codex/config.toml` in a trusted project for project-only setup): ```toml [mcp_servers.lacuna-research] command = "uvx" args = ["lacuna-research-mcp"] ``` Run `codex mcp list` to verify the server is configured. The Codex app, CLI, and IDE extension share this configuration on the same machine. ### Claude Code Add the server for all of your projects with the Claude Code CLI: ```bash claude mcp add --scope user lacuna-research -- uvx lacuna-research-mcp ``` Omit `--scope user` to add it only to the current project. Alternatively, add the following under the top-level `mcpServers` object in `~/.claude.json`: ```json { "mcpServers": { "lacuna-research": { "type": "stdio", "command": "uvx", "args": ["lacuna-research-mcp"] } } } ``` Run `claude mcp get lacuna-research` to verify the server is configured. ### Claude Desktop Open **Settings → Developer → Edit Config**, then add the server under `mcpServers` in `claude_desktop_config.json`: ```json { "mcpServers": { "lacuna-research": { "command": "uvx", "args": ["lacuna-research-mcp"] } } } ``` Restart Claude Desktop after saving the file. ### Other MCP clients For any client that supports local stdio MCP servers, use this standard configuration: ```json { "mcpServers": { "lacuna-research": { "command": "uvx", "args": ["lacuna-research-mcp"] } } } ``` ### Standalone command Install the MCP server as a persistent command: ```bash uv tool install lacuna-research-mcp lacuna-research-mcp ``` Run it without installing a persistent command: ```bash uvx lacuna-research-mcp ``` ### Latest development version PyPI contains tagged releases. To try the latest code from the `main` branch instead: ```bash uvx --from git+https://github.com/tiptreesystems/lacuna-research-mcp.git lacuna-research-mcp ``` ### Local development With uv: ```bash git clone https://github.com/tiptreesystems/lacuna-research-mcp.git cd lacuna-research-mcp uv sync --extra dev uv run lacuna-research-mcp ``` With pip: ```bash cd python3 -m venv .venv . .venv/bin/activate python -m pip install -U pip python -m pip install -e . ``` ## First use After connecting the server, call: 1. `search_lacuna(query="LLM jailbreak defense", search_type="hypothesis", limit=10)` 2. `search_lacuna(query="methods for detecting prompt injection attacks", search_type="works", limit=10)` (production lexical+semantic paper ranking by default) 3. `get_hypothesis(hypothesis_id_or_url="bd35de182c2325ae")` 4. `get_work(work_id_or_url="")` (defaults to `view="context"`) 5. `get_direction(cluster_id_or_url=25108)` (defaults to `view="context"`) ## Scope The corpus covers machine learning and AI research: Works, papers, research directions, authors' research output, venues, institutions, and generated research hypotheses. It does not contain biographies, news, or non-research web content. Agents should answer questions outside that scope from other sources. ## What it exposes - `search_lacuna` Uses Lacuna's public `/api/v1/search` endpoint for directions, Works, papers, authors, venues, institutions, and hypotheses. Explicit Work or paper searches (`search_type="work"` or `"paper"`) use the server's production lexical+semantic ranker when the other ranking arguments remain at their defaults. Pass `search_type="hypothesis"` (or `"hypotheses"` / `"proposal"` / `"proposals"`) for hypothesis search. - `get_hypothesis(hypothesis_id_or_url, view="context")` Hypothesis/proposal. `view="context"` (default) is a compact single-fetch proposal context (summary, abstract, linked directions); `view="full"` returns the server's version record with version history and signal counts. Proposal bodies are in `versions[].markdown`. - `get_direction(cluster_id_or_url, view="context")` Research direction/cluster. `view="context"` (default) requests the compact agent-oriented summary; `view="full"` returns the raw cluster record. - `get_direction_papers(cluster_id_or_url, page, limit, view="compact")` Paginated papers attached to a direction. `view="compact"` (default) returns citation-ready rows (id, url, title, year, venue, a few authors, abstract snippet); `view="full"` returns the raw upstream paper records. - `get_work(work_id_or_url, view="context", figure_limit=None, include_resources=True)` Work lookup for research grouped across versions. Returns the selected version’s content and a version list. `view="context"` is compact; `view="full"` includes concepts, related papers, all figures, and the selected paper record. Public code repositories across all versions are included by default; pass `include_resources=False` to omit them. Use `get_paper` with a version’s `artifact_id` for that paper. - `get_paper(artifact_id_or_url, view="context", figure_limit=None, include_resources=True)` Specific paper-version lookup. Public code repositories are included by default in context and full views; pass `include_resources=False` to omit them. Other views ignore this option. `view="context"` (default) requests the compact agent-oriented context; other views are `"full"`, `"preview"`, `"blog"`, `"figures"`, `"concepts"`, or `"neighbors"`. In context view, `figure_limit` caps the figure preview (server default 3; pass 0 to suppress previews while keeping a `figures_truncated` signal). - Author tools: `get_author_context(…, view="context")`, `get_author_papers`, `get_author_directions`, and `get_author_neighbors`. Start with `get_author_context`, which defaults to the compact agent-oriented view (capped papers plus a readable `impact_directions` list instead of raw `impact_clusters` telemetry). Use the dedicated papers, directions, and neighbors tools to page through those collections without repeating the author context. `view="full"` returns the server-bounded full-shape context (collections remain capped at 100). Pass `include_neighbors=true` to explicitly include similar authors; this may add significant server latency. - Venue and institution tools: `get_venue_context(…, view="context")`, `get_institution_context(…, view="context")`, `get_institution_authors`. Context tools default to compact (capped lists, duplicated blocks dropped; venue keeps a recent-activity slice that always includes the requested `year`). Use `get_institution_authors` to page through an institution's complete author list. ## Wrapped APIs | MCP tool | Lacuna API endpoint | | --- | --- | | `search_lacuna(query, search_type, limit, offset, author_id_or_url, date_from, date_to, venue, sort, ranking_profile, fields)` | `GET /api/v1/search` (`author_id_or_url` constrains `search_type="paper"` to one author, accepting a raw author ID or Lacuna author page URL. `fields` restricts/weights the text fields used for lexical ranking, e.g. `title^4,abstract`, and selects the experimental lexical ranker; for a default relevance-sorted paper search, this bypasses the production lexical+semantic ranker. Allowed fields are `title`, `abstract`, `summary`, `concepts`, `name`, `top_names`, `venue`, each valid only for the types that carry it — `title`: paper/work/cluster/venue/hypothesis; `abstract`/`summary`/`concepts`: paper/work; `name`: author/institution/venue; `top_names`: cluster/hypothesis; `venue`: paper/work/venue (`search_type="all"` spans all). Weights must satisfy `0 < weight <= 100`. Unknown fields, out-of-range weights, type-incompatible fields, and `fields` combined with `ranking_profile="semantic"` are rejected, since the server would otherwise silently drop, cap, or ignore them.) | | `get_hypothesis(hypothesis_id_or_url, view="context")` | `view="context"` → `GET /api/v1/context/hypothesis/{hypothesis_id}?view=compact`; `view="full"` → `GET /api/v1/hypotheses/{hypothesis_id}` | | `get_direction(cluster_id_or_url, view="context")` | `view="context"` → `GET /api/v1/context/direction/{cluster_id}?view=compact`; `view="full"` → `GET /api/v1/clusters/{cluster_id}` | | `get_direction_papers(cluster_id_or_url, page, limit, view="compact")` | `GET /api/v1/clusters/{cluster_id}/papers?view=compact` (default) or `?view=complete` | | `get_work(work_id_or_url, view="context", figure_limit=None, include_resources=True)` | `GET /api/v1/context/work/{work_id}`; `view="context"` → `?view=compact` (`&figure_limit=N` when set); `view="full"` → `?view=complete`; both send `include_resources` | | `get_paper(artifact_id_or_url, view="context", figure_limit=None, include_resources=True)` | `view="context"` → `GET /api/v1/context/paper/{artifact_id}?view=compact` (`&figure_limit=N` when set); `view="full"` → `GET /api/v1/papers/{artifact_id}`; `view="preview"` → `…/preview`; `view="blog"` → `…/blog`; `view="figures"` → `…/figures`; `view="concepts"` → `…/concepts`; `view="neighbors"` → `…/neighbors`; context and full send `include_resources` | | `get_author_papers(author_id_or_url, limit=50, offset=0)` | `GET /api/v1/authors/{author_id}/papers` | | `get_author_directions(author_id_or_url, limit=50, offset=0)` | `GET /api/v1/authors/{author_id}/directions` | | `get_author_context(author_id_or_url, view="context", include_neighbors=false)` | `view="context"` → `GET /api/v1/context/author/{author_id}?view=compact`; `view="full"` → `GET /api/v1/context/author/{author_id}` (`include_neighbors=true` explicitly requests similar authors) | | `get_author_neighbors(author_id_or_url, limit=8, offset=0)` | `GET /api/v1/authors/{author_id}/neighbors` | | `get_venue_context(venue_key_or_url, year, view="context")` | `view="context"` → `GET /api/v1/context/venue/{vkey}[/{year}]?view=compact`; `view="full"` → same route without `view` | | `get_institution_context(institution_key_or_url, view="context")` | `view="context"` → `GET /api/v1/context/institution/{ikey}?view=compact`; `view="full"` → same route without `view` | | `get_institution_authors(institution_key_or_url, limit=50, offset=0)` | `GET /api/v1/institutions/{ikey}/authors` | For id-or-URL arguments, pass either the raw id returned by `search_lacuna` or the corresponding Lacuna page URL. The MCP normalizes Lacuna-relative `url` and `*_url` fields to absolute URLs, and it also absolutifies Lacuna links inside fields named `summary_markdown`, `article_markdown`, `markdown`, `content`, or `description`. `get_author_context` is bounded server-side in both views. The default compact view returns a curated briefing; `view="full"` returns the larger complete shape with embedded collections capped at 100. Use `get_author_papers` and `get_author_directions` rather than trying to page embedded context collections. Search requests are capped at 50 results per call, and direction-paper page requests are capped at 100 results per call. In `search_lacuna`, `date_from` and `date_to` are inclusive publication-date bounds. Accepted formats are `YYYY`, `YYYY-MM`, and `YYYY-MM-DD`; for example, `date_from="2020", date_to="2022-03"` includes papers from January 1, 2020 through March 31, 2022. `search_lacuna` exposes these ranking profiles: - `default` / `lexical` The default profile for all searches. With `search_type="paper"` or `"work"`, `sort="relevance"`, and `fields` unset, it uses the server's production lexical+semantic ranker with graceful fallback. The MCP's default `search_type="all"` uses the server's default lexical ranking instead. - `semantic` Use for embedding-based paper retrieval. The semantic query omits the normal lexical ranking leg, but the server can still overlay exact-title lexical matches. Only supported for `work`, `paper`, and `all` searches (only papers have semantic embeddings). - `bm25_title_abstract` / `bm25` Use for lexical matching constrained to title and abstract. Rejected for `author` and `institution` searches (those records have no title or abstract fields). All searches use the server default unless `ranking_profile` is provided. The MCP rejects unsupported profile/type combinations because the server would otherwise fall back to substring search and silently ignore the requested ranking profile. `sort` accepts `relevance` (default), `year_desc`, and `year_asc`. Year sorts cannot be combined with `ranking_profile="semantic"` — the server would silently ignore the sort — so the MCP rejects that combination; for recent-and-relevant queries, keep semantic ranking and constrain recency with `date_from`/`date_to` instead. ## Environment variables - `LACUNA_SITE_URL` Defaults to `https://lacuna.tiptreesystems.com` - `LACUNA_MCP_TIMEOUT` Defaults to `30` - `LACUNA_MCP_MAX_RETRIES` Defaults to `2`; applies to timeouts, transport errors, and HTTP `429`, `502`, `503`, and `504` responses. - `LACUNA_MCP_USER_AGENT` Defaults to `lacuna-research-mcp/{package_version}` - `LACUNA_MCP_BEARER_TOKEN` Optional bearer token sent as `Authorization: Bearer ...` for private Lacuna deployments. - `LACUNA_MCP_LOG_LEVEL` Defaults to `WARNING` (one of `DEBUG`, `INFO`, `WARNING`, `ERROR`, `CRITICAL`). The default keeps normal operation quiet; lower it only for debugging, since `INFO`/`DEBUG` let the HTTP client log full request URLs — including the search query string — to stderr, which some MCP hosts retain. Environment variables are read once when the MCP server is created, or on the first direct tool/API call if the module is imported without calling `create_mcp()`. ## Implementation layout The server is a thin MCP adapter over Lacuna's HTTP API. The implementation is split by responsibility: - `lacuna_research_mcp/server.py` MCPServer app creation, tool registration, lifespan cleanup, and the `lacuna-research-mcp` CLI entrypoint. - `lacuna_research_mcp/tools.py` MCP tool functions. Each tool normalizes its inputs, calls the matching Lacuna API endpoint through the shared client helpers, and returns JSON-compatible data. - `lacuna_research_mcp/client.py` Runtime HTTP access to Lacuna: the event-loop-bound `httpx.AsyncClient`, retry handling, error wrapping, JSON parsing, URL normalization on responses, and `api_get`/`api_object`/`api_payload`. - `lacuna_research_mcp/config.py` Runtime constants, `RuntimeConfig`, package user-agent construction, and environment parsing. - `lacuna_research_mcp/ids.py` Helpers that accept either raw ids or Lacuna page URLs and extract safe API path segments. - `lacuna_research_mcp/normalize.py` Response post-processing for relative Lacuna URLs and markdown links. - `lacuna_research_mcp/errors.py` User-facing exception type for Lacuna API access failures. ## Notes - `get_paper` and `get_direction` default to `view="context"`. These context views request Lacuna's compact agent-oriented payloads by default to keep MCP responses small. Paper context includes `summary_markdown` when available (otherwise `abstract`), authors, and figures; direction context includes `summary_markdown`, capped papers/authors/related directions, and truncation markers. Use `view="full"` when you need the raw metadata, and the other paper views (`preview`, `blog`, `figures`, `concepts`, `neighbors`) when you want one isolated sub-resource. - An explicit relevance-sorted paper search with no custom `fields` defaults to the server's production lexical+semantic ranker. Set `ranking_profile="semantic"` for embedding-based retrieval (with a possible exact-title overlay) or `"bm25_title_abstract"` for title-and-abstract lexical matching. - Search type aliases are normalized client-side, so `works`, `papers`, `directions`, and `hypotheses` are accepted and mapped to the server's singular values. - Most detail tools accept either the id returned by search or the corresponding Lacuna URL. - Relative Lacuna URLs in `url`/`*_url` response fields and fields named `summary_markdown`, `article_markdown`, `markdown`, `content`, or `description` are normalized to absolute URLs. - Venue and institution keys are opaque hashes (for example `d7bf22905bd6`), never human-readable names like `icml`. Find the key with `search_lacuna(search_type="venue")` first, or pass a `/venue/...` page URL. ## Citation If you find our work helpful, feel free to cite the papers behind Lacuna's research-proposal generation and research map. **Research-proposal generation — Alien Science** ```bibtex @inproceedings{artiles2026alien, title = {Alien Science: Sampling Coherent but Cognitively Unavailable Research Directions from Idea Atoms}, author = {Artiles, Alejandro H. and Weiss, Martin and Brinkmann, Levin and Goyal, Anirudh and Rahaman, Nasim}, booktitle = {ICLR 2026 Workshop on Post-AGI Science and Society}, year = {2026}, url = {https://openreview.net/forum?id=XZWkDET1ia} } ``` **Research map — Lacuna** ```bibtex @misc{weiss2026lacunaresearchmapmachine, title = {Lacuna: A Research Map for Machine Learning}, author = {Martin Weiss and Miles Q. Li and Alejandro H. Artiles and Yacine Mkhinini and Chris Pal and Hugo Larochelle and Nasim Rahaman}, year = {2026}, eprint = {2606.26246}, archivePrefix = {arXiv}, primaryClass = {cs.DL}, url = {https://arxiv.org/abs/2606.26246} } ``` ## License MIT. See [LICENSE](https://github.com/tiptreesystems/lacuna-research-mcp/blob/main/LICENSE).