{ "openapi": "3.0.1", "info": { "title": "Semantic Scholar Paper Search", "description": "Search and extract academic research papers from Semantic Scholar's database of over 200 million publications.", "version": "2.0", "x-build-id": "iEgMJjST3rZfofbe9" }, "servers": [ { "url": "https://api.apify.com/v2" } ], "paths": { "/acts/ryanclinton~semantic-scholar-search/run-sync-get-dataset-items": { "post": { "operationId": "run-sync-get-dataset-items-ryanclinton-semantic-scholar-search", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for its completion, and returns Actor's dataset items in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } }, "/acts/ryanclinton~semantic-scholar-search/runs": { "post": { "operationId": "runs-sync-ryanclinton-semantic-scholar-search", "x-openai-isConsequential": false, "summary": "Executes an Actor and returns information about the initiated run in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK", "content": { "application/json": { "schema": { "$ref": "#/components/schemas/runsResponseSchema" } } } } } } }, "/acts/ryanclinton~semantic-scholar-search/run-sync": { "post": { "operationId": "run-sync-ryanclinton-semantic-scholar-search", "x-openai-isConsequential": false, "summary": "Executes an Actor, waits for completion, and returns the OUTPUT from Key-value store in response.", "tags": [ "Run Actor" ], "requestBody": { "required": true, "content": { "application/json": { "schema": { "$ref": "#/components/schemas/inputSchema" } } } }, "parameters": [ { "name": "token", "in": "query", "required": true, "schema": { "type": "string" }, "description": "Enter your Apify token here" } ], "responses": { "200": { "description": "OK" } } } } }, "components": { "schemas": { "inputSchema": { "type": "object", "properties": { "mode": { "title": "Mode", "enum": [ "auto", "search", "literature-review", "find-foundational", "emerging-trends", "compare-topics", "deep-analysis", "one-answer", "similar-to-papers", "author-papers", "batch-lookup", "snippet-search", "citation-graph" ], "type": "string", "description": "What job you want done. `auto` (default) picks the best mode from your input shape. The other options name the job directly. **Workflow modes** (search-style + opinionated output): `search` keyword search, `literature-review` topic clusters + timeline + ranked picks, `find-foundational` old + still-influential papers (yearTo=now−8, sort by influentialCitationCount), `emerging-trends` recent + fast-rising papers (yearFrom=now−2, sort by velocity). **Graph modes**: `similar-to-papers` runs the Semantic Scholar recommendation engine, `citation-graph` walks citations + references from a seed. **Lookup modes**: `author-papers` pulls every paper an author published, `batch-lookup` resolves up to 500 papers per call by DOI / arXiv ID / PubMed ID, `snippet-search` returns matching passages from open-access PDFs.", "default": "auto" }, "collection": { "title": "Collection (preset)", "enum": [ "top-ml-papers", "recent-ai-breakthroughs", "foundational-deep-learning", "highly-cited-biology", "oncology-recent", "climate-and-energy", "large-language-models", "reinforcement-learning" ], "type": "string", "description": "Optional. Apply a curated query + filter pack so you don't have to configure each field manually. Available: `top-ml-papers` (last 5y, ≥200 cites), `recent-ai-breakthroughs` (last 18mo, high influential), `foundational-deep-learning` (2010-2018 canon), `highly-cited-biology`, `oncology-recent`, `climate-and-energy`, `large-language-models`, `reinforcement-learning`. Your manually-set fields always win — collection only fills gaps." }, "query": { "title": "Search Query", "type": "string", "description": "Used by `search` and (as a fallback) `snippet-search` modes. Search by paper title and abstract (e.g. 'large language models', 'protein folding').", "default": "large language models" }, "yearFrom": { "title": "From Year", "type": "integer", "description": "Earliest publication year. Search-mode only.", "default": 2023 }, "yearTo": { "title": "To Year", "type": "integer", "description": "Latest publication year. Search-mode only." }, "venue": { "title": "Venue", "type": "string", "description": "Filter by journal or conference (e.g. 'Nature', 'NeurIPS', 'ICML', 'ArXiv'). Search-mode only." }, "fieldsOfStudy": { "title": "Field of Study", "enum": [ "Computer Science", "Medicine", "Biology", "Physics", "Chemistry", "Mathematics", "Engineering", "Economics", "Psychology", "Sociology" ], "type": "string", "description": "Filter by academic field. Search-mode only." }, "openAccessOnly": { "title": "Open Access Only", "type": "boolean", "description": "Only return papers with free PDFs available. Search-mode only.", "default": false }, "minCitations": { "title": "Min Citations", "type": "integer", "description": "Minimum citation count. Search-mode only." }, "minInfluentialCitations": { "title": "Min Influential Citations", "type": "integer", "description": "Semantic Scholar's ML-derived signal for citations that meaningfully build on a paper, not just casually mention it. Stricter than raw citation count and unique to S2. Search-mode only." }, "sortBy": { "title": "Sort By", "enum": [ "relevance", "citationCount", "influentialCitationCount", "publicationDate" ], "type": "string", "description": "How to sort results. `influentialCitationCount` is Semantic Scholar's signature ranking — papers cited *meaningfully*, not just frequently. Search-mode only.", "default": "relevance" }, "seedPaperIds": { "title": "Seed Paper IDs", "maxItems": 50, "type": "array", "description": "One or more paper IDs to seed the recommendation engine. Accepts Semantic Scholar paperIds, DOIs, ARXIV:..., or PMID:... — the same identifiers Semantic Scholar resolves. Used by `similar-to-papers` mode. With multiple seeds, the actor switches to multi-paper recommendation (positive + negative blending).", "items": { "type": "string" } }, "negativePaperIds": { "title": "Negative Paper IDs", "maxItems": 50, "type": "array", "description": "Optional. Papers to push the recommendation engine *away* from — useful when you want results similar to A and B but unlike C. Multi-paper mode only.", "items": { "type": "string" } }, "authorName": { "title": "Author Name", "type": "string", "description": "Author's full name (e.g. 'Yann LeCun', 'Geoffrey Hinton'). The actor resolves this against Semantic Scholar's author index and picks the highest-paperCount match. Used by `author-papers` mode." }, "authorId": { "title": "Author ID", "type": "string", "description": "Optional. Semantic Scholar author ID — bypasses name resolution. Find it by inspecting an `authorIds` value from a previous run." }, "paperIds": { "title": "Paper IDs", "maxItems": 2000, "type": "array", "description": "List of identifiers to look up. Up to 500 per Semantic Scholar batch call; the actor chunks longer lists across multiple calls (up to 2,000 total). Accepts: bare DOI, ARXIV:..., PMID:..., or Semantic Scholar paperId. Used by `batch-lookup` mode.", "items": { "type": "string" } }, "snippetQuery": { "title": "Snippet Query", "type": "string", "description": "Text to find inside open-access PDFs. Returns the actual matching passage (~500-char window) with section labels — not just an abstract. Used by `snippet-search` mode. Falls back to `query` if blank." }, "seedPaperId": { "title": "Seed Paper ID", "type": "string", "description": "Single paper ID to walk the citation graph from. Same ID forms as `seedPaperIds`. Used by `citation-graph` mode." }, "direction": { "title": "Direction", "enum": [ "citations", "references", "both" ], "type": "string", "description": "`citations` = papers that cite the seed (forward). `references` = papers the seed cites (backward). `both` splits the budget. Citation-graph-mode only.", "default": "both" }, "complexityLevel": { "title": "Complexity Level", "enum": [ "low", "medium", "high" ], "type": "string", "description": "Controls how much output gets pushed to the dataset. `low` — only `one-answer`, `top-picks`, and `summary` records (compact, exec-friendly). `medium` — adds `recommended-actions`, `research-map`, `blindspots`, `search-insights`, `analysis-pack`. `high` (default) — full dataset with every analytical record. Use `low` when piping into Slack / agent tool calls / dashboards. Use `high` when running interactively or feeding a downstream analytics pipeline.", "default": "high" }, "continuousMode": { "title": "Continuous Mode (deltas only)", "type": "boolean", "description": "Opt-in for scheduled monitoring runs. When `true` AND `monitoringStateKey` is set AND not the first run, only papers that are `isNew=true` OR have `citationDelta>=5` will be pushed. Analytical records (top-picks, research-map, etc.) still emit. Slashes per-run cost on monitoring schedules where most papers haven't changed.", "default": false }, "outputFormat": { "title": "Output Format", "enum": [ "standard", "analysis-pack" ], "type": "string", "description": "`standard` (default) — emit all record types as configured. `analysis-pack` — also emit a single ready-to-use `analysis-pack` record at the end with summary + topFindings + keyPapers + trendNarrative + risks + recommendedActions, suitable for pasting directly into a report or LLM prompt without post-processing. The standard records are still emitted alongside.", "default": "standard" }, "userIntent": { "title": "User Intent (scoring persona)", "enum": [ "researcher", "engineer", "investor", "student" ], "type": "string", "description": "Optional. Switches the impactScore weighting to match your role. `researcher` (default) — citations 55% + influential 25% + velocity 20%. `engineer` — velocity 45% + citations 30% + influential 25% (recency-heavy). `investor` — velocity 50% + citations 25% + influential 25% (momentum-focused). `student` — citations 50% + influential 35% + velocity 15% (foundational-leaning). The active weights are surfaced in logs and on every paper record." }, "queries": { "title": "Queries (compare-topics)", "maxItems": 8, "type": "array", "description": "List of 2–8 query strings to compare side-by-side. Used by `compare-topics` mode. Each query runs a separate search and the actor emits a single `topic-comparison` record showing avg impact, breakout count, foundational count, top paper, and growth-vs-last-run per topic.", "items": { "type": "string" } }, "userProfileKey": { "title": "User Profile Key", "type": "string", "description": "Optional. Set to any string (e.g. `my-research-focus`) to enable lightweight cross-run personalisation. The actor stores the queries you've searched, clusters that have appeared in your runs, and papers that have been returned (FIFO-bounded: 200 queries / 500 clusters / 5,000 papers). Subsequent runs bias the top-picks ranking toward papers in known clusters (up to +20 impact-equivalent boost based on engagement frequency). The summary record carries a `userProfile` aggregate. Distinct from `monitoringStateKey` — monitoring tracks paper deltas; user profile tracks YOUR engagement." }, "monitoringStateKey": { "title": "Monitoring State Key", "type": "string", "description": "Optional. Set to any string (e.g. `weekly-llm-watch`) to turn on cross-run monitoring. The actor stores a snapshot of the papers it returned in a named KV store and on the next run with the same key, every paper record carries `isNew` (true if not seen before), `citationDelta` (citation count change since last run), `previousCitationCount`, and `daysSinceLastSeen`. The summary record carries an aggregate `monitoring` object with new/growing paper counts. First run with a new key sees all `isNew=null` (no prior state). Snapshot capped at 5,000 papers, FIFO. Use one key per scheduled job." }, "apiKey": { "title": "Semantic Scholar API Key", "type": "string", "description": "Optional API key for higher rate limits. Request one at https://www.semanticscholar.org/product/api — unauthenticated users share a small global pool and will hit 429 errors faster on large jobs." }, "maxResults": { "title": "Max Results", "minimum": 1, "maximum": 1000, "type": "integer", "description": "Maximum number of records to return. Caps: search 1000, recommendations 500, author-papers 1000, batch-lookup 500, snippet-search 100, citation-graph 1000.", "default": 50 } } }, "runsResponseSchema": { "type": "object", "properties": { "data": { "type": "object", "properties": { "id": { "type": "string" }, "actId": { "type": "string" }, "userId": { "type": "string" }, "startedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "finishedAt": { "type": "string", "format": "date-time", "example": "2025-01-08T00:00:00.000Z" }, "status": { "type": "string", "example": "READY" }, "meta": { "type": "object", "properties": { "origin": { "type": "string", "example": "API" }, "userAgent": { "type": "string" } } }, "stats": { "type": "object", "properties": { "inputBodyLen": { "type": "integer", "example": 2000 }, "rebootCount": { "type": "integer", "example": 0 }, "restartCount": { "type": "integer", "example": 0 }, "resurrectCount": { "type": "integer", "example": 0 }, "computeUnits": { "type": "integer", "example": 0 } } }, "options": { "type": "object", "properties": { "build": { "type": "string", "example": "latest" }, "timeoutSecs": { "type": "integer", "example": 300 }, "memoryMbytes": { "type": "integer", "example": 1024 }, "diskMbytes": { "type": "integer", "example": 2048 } } }, "buildId": { "type": "string" }, "defaultKeyValueStoreId": { "type": "string" }, "defaultDatasetId": { "type": "string" }, "defaultRequestQueueId": { "type": "string" }, "buildNumber": { "type": "string", "example": "1.0.0" }, "containerUrl": { "type": "string" }, "usage": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "integer", "example": 1 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 } } }, "usageTotalUsd": { "type": "number", "example": 0.00005 }, "usageUsd": { "type": "object", "properties": { "ACTOR_COMPUTE_UNITS": { "type": "integer", "example": 0 }, "DATASET_READS": { "type": "integer", "example": 0 }, "DATASET_WRITES": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_READS": { "type": "integer", "example": 0 }, "KEY_VALUE_STORE_WRITES": { "type": "number", "example": 0.00005 }, "KEY_VALUE_STORE_LISTS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_READS": { "type": "integer", "example": 0 }, "REQUEST_QUEUE_WRITES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_INTERNAL_GBYTES": { "type": "integer", "example": 0 }, "DATA_TRANSFER_EXTERNAL_GBYTES": { "type": "integer", "example": 0 }, "PROXY_RESIDENTIAL_TRANSFER_GBYTES": { "type": "integer", "example": 0 }, "PROXY_SERPS": { "type": "integer", "example": 0 } } } } } } } } } }