skills

Fetch and atomically save verified academic-paper PDFs from legal open-access sources using a DOI or exact title, with access/license provenance and an adjacent hash manifest. Use for automatic OA retrieval or for a publisher URL that must first be resolved to a DOI and may then require an explicitly selected Chrome or optional CloakBrowser user-authorized browser handoff; supports an injectable transport for embedding in research systems.

Fetch full text for a Tiangong KB course document through the Tiangong AI CLI. Use when a task has a course document_id and tags and needs the complete processed text.

Search Tiangong knowledge-base course sources through the Tiangong AI CLI. Use for courseware, lessons, teaching material, Qinghua/Tsinghua humanities archive course collections, and evidence-grounded education-history retrieval. This skill searches only the course source.

Search Tiangong knowledge-base general education sources through the Tiangong AI CLI. Use for broad education knowledge sources only. This skill searches only the edu source, not course or textbook.

Search Tiangong knowledge-base ESG disclosures through the Tiangong AI CLI. Use for environmental, social, and governance reports and disclosure evidence only. This skill searches only the ESG source.

Upload local files or folders into Tiangong KB through the Tiangong AI CLI. Use when a user asks an agent to ingest documents into a KB collection, list uploadable collections, or explain ingest duplicate/status results.

Search Tiangong knowledge-base patent sources through the Tiangong AI CLI. Use for inventions, technical routes, claims, patent evidence, and prior art. This skill searches only the patent source, not sci or report.

Search Tiangong knowledge-base report sources through the Tiangong AI CLI. Use for industry reports, policy reports, whitepapers, institutional reports, and technical reports. This skill searches only the report source, not sci or patent.

Search Tiangong knowledge-base SCI sources through the Tiangong AI CLI. Use for academic papers, scientific journal evidence, literature support, and research claims. This skill searches only the sci source, not report or patent.

Search Tiangong knowledge-base textbook sources through the Tiangong AI CLI. Use for textbook-style explanations and reference material only. This skill searches only the textbook source, not course or edu.

Upload local documents to TianGong AI Unstructure `/mineru_with_images` API for fine-grained parsing and return only plain fulltext content. Use when a task needs document fulltext extraction with `return_txt=true`, strict file-type allowlist validation, API base URL/auth token from environment variables, and optional provider/model overrides.

Help users build an initial mental model of an unfamiliar domain when they do not know where to start, do not know what questions to ask, or lack the vocabulary, structure, and learning path needed to explore the field effectively.

Query GDELT DOC 2.0 API using explicit query/mode/format/time parameters with retry, throttling, and detailed logs. Use when tasks need topical/domain retrieval, article lists, or timeline aggregates without pulling raw events files.

Fetch USGS Water Services Instantaneous Values JSON for one bounding box or explicit site list, then return structured hydrology observations with site metadata, parameter codes, timestamps, qualifiers, and validation output. Use when tasks need deterministic station-based streamflow or gage-height evidence from USGS for flood, runoff, river, or water-level verification in the United States.

Fetch AirNow Hourly AQ Obs file products from official AirNow file endpoints and return structured site-hour-parameter records filtered by UTC time window, bounding box, and pollutant parameter list. Use when tasks need deterministic AirNow monitoring-site observations at hourly cadence, especially for broad geography/time extraction where AirNow file products are preferred over per-request web-service loops.

Fetch Federal Register document search results with explicit term, publication-window, agency, document-type, topic, docket, and RIN filters, then return structured official-policy records with retries, throttling, pagination, and validation. Use when tasks need authoritative U.S. Federal Register notices, proposed rules, final rules, or presidential documents for policy verification and regulatory context.

Analyze river outfall Excel workbooks and build report-ready river outfall status visualizations that compare current, normal, 20-year, and 50-year water levels, distinguish left-bank and right-bank outfalls, and identify which outfalls are safe, partially submerged, or fully submerged. Use when Codex needs to turn a river longitudinal-profile or outfall inventory workbook into a briefing chart, status summary, mock dataset, or reusable HTML report.

Fetch Bluesky seed posts and expand reply-thread cascades in configurable UTC time windows with retries, throttling, transport checks, and structure validation. Use when tasks need social-opinion diffusion signals from Bluesky (query/feed/list/author sources) as JSON/JSONL artifacts for downstream sentiment or viral-cascade analysis.

Fetch Open-Meteo Historical Weather API time-window data for weather and shallow-soil variables with retries, throttling, detailed logs, and transport/structure validation. Use when tasks need deterministic physical-world context for ecology or environmental verification, such as checking historical temperature, precipitation, wind, humidity, FAO evapotranspiration, soil temperature, or soil moisture for one or more coordinates in a specified date range.

Fetch YouTube public comment threads and replies for discovered or provided video IDs in configurable UTC time windows with retries, throttling, transport checks, and structure validation. Use when tasks need YouTube public-language datasets after domain video discovery, especially as a complement to official-language sources such as GDELT or Regulations.gov.

Search YouTube videos for topical/domain discovery using explicit query strings, channel filters, publish-time windows, retries, throttling, enrichment, and validation. Use when tasks need candidate YouTube videos for a domain such as environment/public-opinion research before downstream comment collection or when OpenClaw must decide search keywords itself from a broad task objective.

Fetch NASA FIRMS active fire detections from the area/csv endpoint for specified date windows, sources, and bounding boxes with retries, throttling, detailed logs, and transport/structure validation. Use when tasks need deterministic wildfire, open-burning, or smoke-source verification for ecology and environmental monitoring within a known coordinate window.

Fetch Open-Meteo Air Quality API hourly background fields for specified date windows and coordinates with retries, throttling, detailed logs, and transport/structure validation. Use when tasks need deterministic modeled air-quality context for ecology or environmental verification, such as checking PM, gases, dust, UV, pollen, or AQI background conditions beyond station coverage.

Fetch Open-Meteo Flood API daily river-discharge data for specified date windows and coordinates with retries, throttling, detailed logs, and transport/structure validation. Use when tasks need deterministic flood or runoff verification for ecology and environmental monitoring, such as checking river discharge background, flood risk windows, or ensemble-style discharge signals for one or more coordinates.

Upload local files to a Dify pipeline knowledge base through the published pipeline datasource flow and optionally apply existing metadata fields. Use when the target dataset is a published rag_pipeline/local_file knowledge base or ordinary Dify document create APIs return completed with 0 chunks.

Analyze local image files with ImageHash and OpenCV to detect near-duplicate photos and blurry shots, then preview or remove them. Use when Codex needs to clean a photo folder, deduplicate similar images, identify blur, or safely delete/move bad images after reviewing a report.

Fetch GDELT 2.0 Events export snapshots from lastupdate/masterfilelist with retry, throttling, and detailed logs. Use when tasks need latest or time-range event data files (.export.CSV.zip) for deterministic ingestion and analysis.

Fetch GDELT 2.0 GKG export snapshots from lastupdate/masterfilelist with retry, throttling, transport validation, and structure validation. Use when tasks need latest or time-range GKG files (*.gkg.csv.zip) for deterministic ingestion and machine-readable manifests.

Fetch GDELT 2.0 Global Mentions export snapshots from lastupdate/masterfilelist with retry, throttling, transport validation, and structure validation. Use when tasks need latest or time-range mentions files (*.mentions.CSV.zip) for deterministic ingestion and machine-readable manifests.

Fetch Regulations.gov v4 comment detail resources by comment IDs with authenticated requests, retries, throttling, transport checks, and structure validation. Use when tasks need enriched comment payloads (full comment text, docket/document linkage, and optional attachment relationships) after a comments list fetch.

Fetch Regulations.gov v4 comments in configurable time windows with authenticated requests, pagination, retries, throttling, transport checks, and response-structure validation. Use when tasks need policy/public-opinion comment datasets for downstream analysis (including sentiment workflows) with deterministic JSON/JSONL outputs.

Convert local image files from common raster formats into JPG while setting a target DPI (default 96). Use when Codex needs to convert a file or folder of images such as HEIC, HEIF, PNG, TIFF, WebP, BMP, GIF, AVIF, or JPEG into `.jpg` for upload, OCR, archiving, or tools that require JPEG output.

Fetch DOI candidates from PostgreSQL `journals` and batch write `abstract` by DOI. Use when you need DB-only backfill workflow: (1) select newest rows where doi is not empty, author is not empty, and abstract is empty (default limit 10), then (2) write prepared abstract values back in batch by DOI.

Retrieve OpenAQ metadata and measurements through authenticated API queries or public S3 archive fetches with automatic source routing. Use when tasks need OpenAQ data discovery across countries/providers/locations/sensors/parameters, filtered current or historical API queries, or large historical backfills from the AWS Open Data archive.

Analyze one GitHub user's code contribution activity in a specified UTC time window and generate a Markdown report focused on practical contributions (authored merged PRs, PRs merged by the target user, and authored commits) across accessible public and private repositories. Use when users ask to evaluate real engineering contribution over a period and want evidence links.

Download Figshare files via browser-only workflow (no web_fetch/curl probing). Use when a user provides a Figshare DOI or ndownloader URL and needs a reliable fetch path that always opens the page in browser, triggers the real download, then copies the downloaded file into a target path.

NotebookLM CLI wrapper via `python3 {baseDir}/scripts/notebooklm.py` (backed by notebooklm-py). Use for auth, notebooks, chat, sources, notes, sharing, research, and artifact generation/download.

Operate Synology DSM File Station via WebAPI for major file workflows including listing, search, folder creation, rename, copy/move, delete, upload/download, and archive extract. Use when tasks need scripted NAS file operations with service address, username, and password loaded from environment variables. Note: compress is temporarily unavailable in this skill.

Fetch recent `journals` rows directly from PostgreSQL and return a JSON array containing `doi`, `title`, `abstract`, and `date` (filtered by rolling `DAYS` on `created_at`). Use when users need raw paper payloads for downstream processing without LLM filtering or hotspot report generation.

Fetch journal articles from Crossref published after a user-specified date and insert them into PostgreSQL `journals` with DOI deduplication. Use when incrementally ingesting journal metadata from `journals_issn` into `journals`.