Automation or dependency surfaces exist, but .github/dependabot.yml is missing.
Add Dependabot updates for GitHub Actions and dependency manifests.
Dow Theory MCP Server: 100+ Deterministic Tools, Skill & RAG by PDFtoAgent
babaksh/pdftoagent-hamilton · v1.0.0 · Tools & Integrations
If you've fed technical books into standard RAG, you know the flaw: "LLMs hallucinate math".
When evaluating quantitative rules, models guess tokens instead of calculating exact logic.
Guard checks the skills and MCP servers a plugin adds on your machine and asks you before a risky action runs. The scores above are free, and no plan changes them.
From the plugin's manifest and HOL's scan. Anything the plugin does not declare is shown as not declared.
💡 The Solution: Triad Architecture
PDFtoAgent turns domain literature into a unified FastMCP Server:
1. 📚 Vector RAG: Semantic search with exact page and image citations.
2. 🧠 Codified Skill (SKILL.md): Procedural heuristics and decision trees.
3. ⚙️ Deterministic Tools: Verified Python algorithms with 0% calculation errors.
🏛️ The Proof: 1922 Dow Theory Showcase
I turned William Peter Hamilton’s 1922 classic, "The Stock Market Barometer", into 112 deterministic FastMCP tools + Skill + RAG in one package!
Ask the agent if a pullback to 91 after a rise from 80 to 100 is a valid secondary reaction:
Available as a 1-click Zero-Docker Windows Portable package (run.bat) and Docker image. Ready for Claude Desktop, Cursor, and any MCP client!
Listed in 0 of 5 directory lists
Read from the list each directory publishes on GitHub. Being listed means the directory accepted the plugin; it is not a HOL security rating.
Plugins Anthropic maintains or selected for the marketplace Claude Code ships with.
Plugins submitted through claude.ai that passed Anthropic’s review; the list is a nightly mirror.
Plugins OpenAI lists in the marketplace Codex ships with.
Plugins Cursor publishes itself. Cursor does not publish its list of third-party Marketplace plugins.
The community Codex plugin list HOL maintains on GitHub.
Lists last read . HOL re-reads them every day.
Submitting it? Check it against each directory's rules
HOL cannot see every install. These are separate signals from different sources, so they are not added together. Counts under each floor show as <N.
No reviews yet.
Read or write reviewsSimilar by contents
Suggested from HOL scan evidence: same category, at least one shared runtime, similar tags and contents. Paid placement and Trust Score play no part.