research-map

Generates concept-to-code maps linking paper equations and algorithms to repository locations.

Updated Jul 3, 2026
One-click install
npx skills add https://github.com/mnazaal/dotfiles --skill research-map-mnazaal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: research-map
Source: https://github.com/mnazaal/dotfiles/tree/main/.agents/skills/research-map
Command: npx skills add https://github.com/mnazaal/dotfiles --skill research-map-mnazaal

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Research projects drift out of sync with their papers: it becomes unclear which code implements which equation, where the implementation diverges from the paper, and how the architecture fits together. This Skill regenerates an on-demand orientation map connecting paper claims to code locations. ## Core Features & Use Cases - Concept-to-Code Mapping: Maps paper equations, algorithms, and concepts to exact code locations (e.g., "Eq. 7 ELBO → losses.py:elbo"). - Architecture and Dataflow Diagrams: Produces small mermaid diagrams of module dependencies and data pipelines. - Divergence Detection: Lists evidence-backed paper-code divergences with file:line and paper-section citations. - Three Output Sinks: Emits to chat for orientation, stamps a README Overview block as the shareable snapshot, or writes a full self-contained notes/main.html with MathJax. - Use Case: Onboarding onto an unfamiliar ML research repo, ask for a project map to see which files implement the paper's training algorithm and where the code silently differs from the published method. ## Quick Start Map this research project by linking each paper equation and algorithm to its code location and listing any divergences.

Frequently Asked Questions about research-map

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I map a research paper to its codebase?▼

Read the paper for its problem, method, and key equations, then scout the repository for entry points and modules. Build a table mapping each equation or algorithm to its code location, such as "Algorithm 1 → train.py:fit", and note divergences with file:line evidence.

How to generate architecture diagrams for a research repository?▼

Create small mermaid diagrams showing module dependencies and dataflow from data through transforms, model, and metrics. Keep diagrams under roughly 12 nodes and split larger systems into multiple diagrams rather than cramming.

Can the project overview be kept in the README automatically?▼

Yes, the summary, mermaid diagram, and concept-to-code table are written between delimited research-map:begin/end markers in the README. Each write carries a provenance stamp with the git commit hash and date so staleness is visible.

What happens when the code changes after a map is generated?▼

Persisted outputs are treated as stale once HEAD moves past the stamped commit and are regenerated rather than trusted. Maps are never hand-edited; they are always derived fresh from the current code and paper.

Does this skill modify source code while mapping?▼

No, it is read-only on code. The only files it writes are the README section between its markers and the notes/main.html file, which is replaced wholesale on regeneration.