Menu
Extract data from PDFs and query CSV files
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Build animejs.com-grade scroll-driven kinetic pages with the vendored anime.js v4 bundle (website/vendor/anime.esm.min.js). Use when working on the prime-silo marketing site (website/) or any scrollytelling/animation surface. Contains the VERIFIED v4 API, the design language to emulate, the Prime-Silo set-piece specs, and the verification protocol.
Operate, analyze, or extend the LONGVIEW session-synthesis pipeline (cards, graph, reports, book, audiobook). Use when working on scripts/longview, reviewing card output, planning card schema changes, or debugging a LONGVIEW run.
Operate the Prime-Silo delivery board (delivery/) autonomously — select, claim, execute, verify, and log work contracts without frontier supervision. Use whenever picking up a task from delivery/board/BOARD.md, authoring a gate script, verifying another agent's task, or deciding which model tier should take which contract.
The Prime-Silo product experience contract — identity (Prime-Silo + Agent Benny, never Space Agent), earth-tone design tokens, progressive-discovery doctrine, and the "motion is meaning" choreography rules that bring animejs.com-grade kinetics INTO the app without violating the C0 calm contract. Use for any user-facing UI work in app/, server/pages/, or packaging/ (login, first-run, shell, mascot, Studio, Bridge).
DRAFT — pick up and deliver one delivery/tasks/ work contract under the six-sigma protocol (deterministic selection, TDD, allowlist discipline, author≠verifier). Use when asked to "take the next task", "work the board", or deliver a specific task id. Full mechanization lands with W1 (work next) and W2 (sandbox provisioning).
Create, patch, inspect, and remove widgets in the current space
| name | data_processing |
| description | Extract data from PDFs and query CSV files |
Use this skill when analyzing tabular data (CSV) or unstructured PDF documents within the workspace. This gives you the ability to query dataframes via Pandas and extract raw text (including OCR fallback via PyMuPDF/Tesseract) from PDFs.
[!WARNING] While
extract_pdf_textis useful for quick, ad-hoc reads, do NOT use this to feed multi-agent swarms or complex analyses for files located in thestaging/directory. Those files must be formally ingested into the Knowledge Graph via arag_ingestpipeline first.
extract_pdf_text(pdf_path, workspace="default") - Extracts all text from a given PDF file in data_in. Supports fallback OCR.query_csv(csv_path, query, workspace="default") - Query a CSV file using Pandas. The query can be a standard string like "amount > 100" or a dataframe expression like "df.describe()".Action: query_csv
Action Input: {"csv_path": "financials.csv", "query": "df.head(5)"}
Observation: Returns the first 5 rows formatted as a markdown table.
Action: extract_pdf_text
Action Input: {"pdf_path": "architecture_diagram.pdf"}
Observation: Returns the text content extracted from the document.