agentleFS
EodHistoricalDataEodHistoricalData/eodhd-claude-skills/CLAUDE.md

eodhd-claude-skills

CLAUDE.md47 starsChanged 4 months agoMITView on GitHub

Plugin enabling AI agents (Claude Code, Codex) to work with the EODHD financial data API. Distributed as a Claude Code plugin (EodHistoricalData/eodhd-claude-skills). Version: 0.5.2. Includes MCP Server connector (OAuth), 13 skills (1 core + 7 workflow + 5 user-facing commands), financial analyst agent, 64 endpoint docs, 28 general guides, and a stdlib-only Python client. As of v0.5.2 the user commands (eodhd-analyze etc.) live under skills/ as standard skills and are invoked namespaced (e.g. /eodhd-api:eodhd-analyze) — the legacy commands/ directory was…

The file

We split this CLAUDE.md into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No runtime deployment. Distributed via GitHub releases.

    line 99
  2. Distinctive:

    Commit style: imperative, descriptive

    line 124

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

EODHD Claude Skills

Overview

Plugin enabling AI agents (Claude Code, Codex) to work with the EODHD financial data API. Distributed as a Claude Code plugin (EodHistoricalData/eodhd-claude-skills). Version: 0.5.2.

Includes MCP Server connector (OAuth), 13 skills (1 core + 7 workflow + 5 user-facing commands), financial analyst agent, 64 endpoint docs, 28 general guides, and a stdlib-only Python client. As of v0.5.2 the user commands (eodhd-analyze etc.) live under skills/ as standard skills and are invoked namespaced (e.g. /eodhd-api:eodhd-analyze) — the legacy commands/ directory was removed because its flat files did not load as usable commands in a live session on newer Claude Code (verified live on 2.1.158 and 2.1.161: all 13 skills load).

File Structure

.claude-plugin/
  marketplace.json              # Plugin manifest (name, version, skills list)
  plugin.json                   # Extended plugin metadata (keywords, author, license)
.mcp.json                       # MCP Server connector → mcp.eodhd.com/v2/mcp
agents/
  financial-analyst.md          # Financial analyst agent definition
skills/                         # All skills incl. user commands (auto-discovered)
  eodhd-analyze/SKILL.md        # /eodhd-api:eodhd-analyze <ticker>
  eodhd-compare/SKILL.md        # /eodhd-api:eodhd-compare <ticker1> <ticker2>
  eodhd-market/SKILL.md         # /eodhd-market
  eodhd-screen/SKILL.md         # /eodhd-screen <criteria>
  eodhd-macro/SKILL.md          # /eodhd-macro
  eodhd-api/                    # Core skill — full API access (30+ client endpoints)
    SKILL.md
    references/
      general/                  # 28 general guides
      endpoints/                # 64 endpoint docs
      subscriptions/            # 7 subscription plans
      workflows.md              # 4 analysis patterns
    scripts/
      eodhd_client.py           # Python client (stdlib-only, ~495 lines)
      market_cap_series.py
      test_investverte_*.py
    templates/
      analysis_report.md
  company-brief/SKILL.md        # Company snapshot workflow
  earnings-monitor/SKILL.md     # Earnings tracking workflow
  market-overview/SKILL.md      # Market summary workflow
  portfolio-risk/SKILL.md       # Risk analysis workflow
  stock-screener/SKILL.md       # Stock screening workflow
  macro-dashboard/SKILL.md      # Macro-economic dashboard
  options-analyzer/SKILL.md     # Options analysis workflow
adapters/
  claude/eodhd-api.md
  codex/eodhd-api.md
registry/
  capabilities.json             # Single source of truth for all endpoints (67 entries)
  build.py                      # Generates support-matrix.md; --check for CI freshness
  README.md                     # Registry schema + workflow

Key Components

ComponentPurposeLocation
marketplace.jsonPlugin manifest (version, skills list).claude-plugin/marketplace.json
plugin.jsonExtended metadata (keywords, author, license).claude-plugin/plugin.json
.mcp.jsonMCP Server connector (OAuth HTTP).mcp.json
Core SKILL.mdFull API skill (triggers, workflow, guardrails)skills/eodhd-api/SKILL.md
Workflow skills (7)Curated analysis workflowsskills/*/SKILL.md
Financial analystAgent definitionagents/financial-analyst.md
Slash commands (5)User-facing commands (standard skills, invoked /eodhd-api:<name>)skills/eodhd-{analyze,compare,market,screen,macro}/SKILL.md
eodhd_client.pyStdlib-only Python client, 31 endpointsskills/eodhd-api/scripts/eodhd_client.py
Endpoint docsPer-endpoint API referenceskills/eodhd-api/references/endpoints/
General guidesCross-cutting topicsskills/eodhd-api/references/general/
Capabilities registryMachine-readable source of truth for endpoints + support tiersregistry/capabilities.json
Support matrixGenerated agent-facing endpoint/tier tableskills/eodhd-api/references/general/support-matrix.md

Development Setup

  1. Clone: git clone https://github.com/EodHistoricalData/eodhd-claude-skills.git
  2. Set token: export EODHD_API_TOKEN="your_token_here"
  3. Test client: python skills/eodhd-api/scripts/eodhd_client.py --endpoint exchanges-list

Or install as Claude Code plugin:

/plugin marketplace add EodHistoricalData/eodhd-claude-skills
/plugin install eodhd-api@eodhd-claude-skills

Testing

python skills/eodhd-api/scripts/eodhd_client.py --endpoint exchanges-list
python skills/eodhd-api/scripts/eodhd_client.py --endpoint eod --symbol AAPL.US --from-date 2025-01-01 --to-date 2025-01-31

Automated tests live in tests/ (stdlib-only, no pytest dependency):

  • tests/test_python_client.py — 30 e2e cases hitting every endpoint in SUPPORTED_ENDPOINTS against the live API. Requires EODHD_API_TOKEN.
  • tests/test_mcp_v1.py — MCP protocol e2e: v2 OAuth challenge, v1 initialize/tools-list/tools-call.
  • tests/test_skill_references.py — static cross-reference validation (SKILL.md frontmatter, endpoint docs, slash commands, agent frontmatter, manifest tool count vs MCP).
  • tests/test_registry.py — static parity: registry ↔ client SUPPORTED_ENDPOINTS, registry ↔ endpoint docs, validated-tier ↔ e2e cases, and support-matrix freshness. Stdlib-only, runs in CI.

CI (.github/workflows/validate.yml) runs the static suite on every push/PR. The two e2e suites run automatically when EODHD_API_TOKEN is configured as a GitHub Actions secret; otherwise the e2e job emits a skip warning and the CI passes (no false negatives).

Deployment

No runtime deployment. Distributed via GitHub releases.

  • Version bumps: edit version in .claude-plugin/marketplace.json and .claude-plugin/plugin.json
  • Auto-release: .github/workflows/release.yml triggers on version bump
  • Users install/update via Claude Code plugin system

Conventions

Documentation

  • Endpoint docs follow standard template: Purpose, Parameters, Response, Examples, Notes
  • exchanges.md = source of truth for exchange codes
  • symbol-format.md = source of truth for ticker formatting

Python Client

  • eodhd_client.py is stdlib-only — no external packages
  • Token from EODHD_API_TOKEN env var
  • JSON stdout, exit codes: 0=success, 1=API error, 2=client error

Skills

  • Each workflow skill has YAML frontmatter (name, description, version)
  • Skills reference endpoints from ../eodhd-api/references/endpoints/
  • Skills include output structure templates and endpoint cost tables

Git

  • Main branch: main
  • Commit style: imperative, descriptive
  • Version in marketplace.json and plugin.json must stay in sync

Related

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.