external-context

Run parallel specialist searches and aggregate evidence summaries with DOI/URL lists.

32|1|Updated May 19, 2026
One-click install
npx skills add https://github.com/woodfishhhh/EZ_math_model --skill external-context-woodfishhhh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: external-context
Source: https://github.com/woodfishhhh/EZ_math_model/tree/main/skills/ez-math-model/tools/external-context
Command: npx skills add https://github.com/woodfishhhh/EZ_math_model --skill external-context-woodfishhhh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of insufficient or single-thread external knowledge when building a math modeling solution, where multiple domain angles (methods, theory, industry context, sensitivity analysis) must be retrieved and cross-validated.

Core Features & Use Cases

  • Parallel specialist external knowledge gathering: Runs 2–3 specialist research queries in multi/hybrid mode to cover different knowledge needs at once.
  • Compact, decision-ready evidence summaries: Produces a short summary capped in length per specialist plus DOI/URL lists for traceable sources.
  • Conflict-aware aggregation: Keeps all findings when specialists disagree and labels the controversial points for later modeler/writer review.
  • Graceful fallback strategy: If the host skill is unavailable or specialists fail, it degrades to sequential retrieval such as paper-search or webcrawl.

Quick Start

Use the external-context skill in multi or hybrid mode to ask several specialists to search for the model's classic literature, the problem domain's industry benchmarks, and sensitivity-analysis methods, then collect the aggregated output in the workdir path workdir/{task_id}/refs/external_context.md.

Frequently Asked Questions about external-context

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

FAQPage Schema
How do I gather external knowledge for math modeling from multiple domain angles?

To gather external knowledge for math modeling, you can run parallel specialist searches that concurrently retrieve domain theory, industry context, and sensitivity-analysis references, aggregating them into a single markdown file with DOI/URL lists.

Can I run multiple literature review searches in parallel for a hybrid math modeling project?

Yes, you can run 2–3 specialist literature review queries concurrently in multi or hybrid mode to cover different knowledge needs at once, rather than relying on a single-thread search query.

How does conflict-aware aggregation handle disagreements between specialist search results?

Conflict-aware aggregation retains all findings when specialist search results disagree, explicitly labeling the controversial points for later modeler or writer review instead of discarding conflicting data.

What happens if the host skill is unavailable during a parallel domain research search?

If the host skill is unavailable or specialist searches fail, the system uses a graceful fallback strategy that degrades to sequential retrieval methods such as paper-search or webcrawl.

Does this parallel research approach provide traceable sources for sensitivity analysis references?

Yes, the approach provides traceable sources by capping each specialist output into a concise summary accompanied by DOI and URL lists, ensuring all sensitivity-analysis references are verifiable.

Where are the aggregated external context results stored for multi-agent math modeling?

The aggregated external context results are stored in the directory path workdir/{task_id}/refs/external_context.md, consolidating all specialist search findings into a single decision-ready file.