Sign InOpen Brain
AI EngineerVideoSource Linked

How We Got LLMs to Recommend Our Open Source Library — Christopher Burns, Inth

For fast-changing libraries, ship concise agent-facing docs inside the package as well as on the web; agents may inspect installed code before visiting your site.

AI Engineer · Aug 26, 2026
Open Source Open MarkdownOpen JSON
Source Summary

C15T reports **3 million npm downloads**, with LLM recommendations now its largest inbound source. Its approach combines a hand-written llms.txt, Markdown variants of pages, WebMCP tools, and bundled package docs referenced by AGENTS.md.

Practical Implication

Start with a concise, manually curated llms.txt, expose page-level Markdown through multiple access paths, and bundle current docs with rapidly changing libraries. The speaker’s testing found roughly **40 useful lines** beat 1,000 noisy ones and package-local docs saved nearly **50% of tokens** across several models.

Agent-Ready Context
C15T reports **3 million npm downloads**, with LLM recommendations now its largest inbound source. Its approach combines a hand-written llms.txt, Markdown variants of pages, WebMCP tools, and bundled package docs referenced by AGENTS.md.

Start with a concise, manually curated llms.txt, expose page-level Markdown through multiple access paths, and bundle current docs with rapidly changing libraries. The speaker’s testing found roughly **40 useful lines** beat 1,000 noisy ones and package-local docs saved nearly **50% of tokens** across several models.

The talk does not isolate which change caused recommendation growth, and support for Markdown headers, URL conventions, WebMCP, and agent-readiness tests remains inconsistent. Treat each mechanism as an experiment and measure it against real agent tasks.
Connected Context · Feed7 Judgment

This extends context engineering from repository use to whether agents can discover, understand, and recommend a library at all. It favors concise curated entry points and package-local, current documentation over exhaustive context, while treating llms.txt, Markdown exposure, and WebMCP as separately measurable distribution experiments. The recommendation growth and token savings remain local evidence because attribution and format support are unresolved.

JetBrains/go-modern-guidelinesJetBrains provides a concrete version-bounded example of bundling current guidance for agents, reinforcing package-local documentation while showing that freshness must also respect each project’s toolchain constraints.How to Generate Mergeable Code with a Context Engine — Peter Werry, UnblockedWerry expands the same retrieval principle beyond public library docs: agents need task-relevant decisions, conventions, and history, not merely plausible code or broadly available documentation.RLM: Recursive Language Models for Large Codebases - Shashi, Superagentic AIRLM’s bounded evidence retrieval supports the finding that compact relevant context can outperform loading large noisy corpora, though it applies that principle to repositories rather than library distribution.Graphify-Labs/graphifyGraphify offers a structured alternative for reducing repeated discovery work inside projects; unlike the library’s static Markdown paths, it preserves queryable relationships and distinguishes inference from extracted facts.
Context Map
contextcoding#context-engineering#retrieval#coding-agents
Uncertainty
The talk does not isolate which change caused recommendation growth, and support for Markdown headers, URL conventions, WebMCP, and agent-readiness tests remains inconsistent. Treat each mechanism as an experiment and measure it against real agent tasks.