lseg-data — Claude Code skill: what it does, how to install
Configures AI coding agents to retrieve LSEG and Refinitiv financial data via the lseg.data Python package or Workspace Web sessions.
no license 1 installs ★ 27 updated 2026-10-29
npx skills add edwinhu/workflows --skill lseg-data -a claude-codeWhat lseg-data does
This skill guides the agent across multiple routes to fetch London Stock Exchange Group (formerly Refinitiv) records. It organizes access into ordered tiers: the primary lseg.data Python package with standard RDP credentials, a token extraction flow via browser CDP for machines lacking dedicated API keys, and internal in-page browser queries for specialized endpoints such as FSCREEN or SDC deal universes.
Silent query failures are a major hazard in Refinitiv endpoints. Misspelled field prefixes or missing market suffixes like .O or .L yield empty dataframes rather than explicit errors. The instructions enforce step-by-step verification, demanding head inspections and null checks before any data summary reaches the user.
The native Workspace desktop client runs only on Windows and macOS. When working in Linux environments, all browser-based flows must rely solely on the web interface at https://workspace.refinitiv.com/web via Chromium remote debugging.
What it helps with
- query corporate fundamentals and ESG scores using lseg.data
- extract Workspace Web tokens over Chrome DevTools Protocol
- verify RIC exchange suffixes to prevent silent query drops
- retrieve deal metrics for IPOs, private equity, and syndicated loans
Example prompts for the agent
- Query annual revenue and fundamentals for AAPL.O using the lseg.data library
- Extract current edp-token from the open Workspace Web tab via scripts/workspace_cdp.py
- Verify field prefixes and sample output rows for our Refinitiv data pull
How to install lseg-data
1Skills CLI (Vercel)
npx skills add edwinhu/workflows --skill lseg-data -a claude-code2Manually: put the files in place
~/.claude/skills/lseg-data/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/lseg-data/
1Skills CLI (Vercel)
npx skills add edwinhu/workflows --skill lseg-data -a codex2Manually: put the files in place
.agents/skills/lseg-data/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add edwinhu/workflows --skill lseg-data -a cursor2Manually: put the files in place
.cursor/skills/lseg-data/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install edwinhu/workflows skills/lseg-data2Manually: put the files in place
.github/skills/lseg-data/SKILL.md1Skills CLI (Vercel)
npx skills add edwinhu/workflows --skill lseg-data -a gemini-cli2Platform CLI
gemini skills install https://github.com/edwinhu/workflows --path skills/lseg-dataManually: put the files in place
.devin/skills/lseg-data/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add edwinhu/workflows --skill lseg-data -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about lseg-data
How to install lseg-data in Claude Code?
Run npx skills add edwinhu/workflows --skill lseg-data -a claude-code. The rule will be installed to ~/.claude/skills/lseg-data/SKILL.md.
What does lseg-data do?
It structures prompts and scripts to query Refinitiv and LSEG data feeds cleanly, handling token extraction and guarding against silent null returns.
Why does the skill bypass the Claude LSEG MCP connector?
Pre-registered MCP connector tools fail with entitlement errors unless specific flags like MCP_DATA_ON are enabled on the enterprise subscription, making local Python calls more dependable.