Research

Research Program

Our research program is organized around a core question: What does the agent ecosystem need to mature into reliable, long-lived infrastructure for software development?

Current Research Areas

Documentation and information access How agents consume documentation, where they fail, and what site owners can do about it. Standards work to codify best practices. This area produced the Agent-Friendly Documentation Spec and the afdocs audit tool. Active testing is measuring what happens between “agent fetches URL” and “model sees content” across platforms including Claude, Cursor, Gemini, and OpenAI, documenting truncation limits, HTML processing, and content negotiation behavior. A documentation ecosystem report is in progress, similar to the Agent Skill Report.

Tool quality and interoperability Evaluating MCP servers, Agent Skills, and other tool integrations for reliability, spec compliance, and developer experience. This area produced the Agent Skill Report and the skill-validator analysis tool (v1.3.0, stable), with an enterprise variant available for AWS Bedrock environments. A new community research project, Agent Skill Implementation, is empirically testing how platforms actually load, manage, and present Agent Skills, with 23 checks across 9 categories and benchmark skills for cross-platform testing. An MCP ecosystem report is in progress, investigating 10 research questions around when MCP servers justify their complexity, including abstraction value over direct API calls, dynamic capability registration, ecosystem portability, security isolation, and registry/discovery gaps.

Ecosystem health signals Developing metrics and methodologies for tracking the overall health of the agent ecosystem over time.

Agent behavior patterns How agents navigate code, use context, handle errors, and interact with external resources in real development workflows.

Standards and advocacy Driving adoption of specifications that make the ecosystem work better for everyone: agents, developers, and platform builders alike.

Published Reports

• Agent Skill Report — Qualitative analysis of 673+ public Agent Skills, including findings on spec compliance issues across the ecosystem.
• Agent Skill Implementation — Community research project empirically testing how 25+ agent platforms load, manage, and present Agent Skills. 23 checks across 9 categories (loading timing, directory recognition, resource access, content presentation, lifecycle management, access control, structural edge cases, skill-to-skill invocation, and skill dependencies), with 17 benchmark skills containing canary phrases for cross-platform testing. Open for community contributions of per-platform findings.
• Docs Ecosystem Report — In progress. Qualitative analysis of how well the documentation industry serves the agent ecosystem, based on targeted audits and agent behavioral evaluations.
• MCP Ecosystem Report — In progress. Investigating when MCP servers earn their complexity: abstraction value, dynamic capabilities, CLI alternatives, ecosystem portability, security isolation, performance overhead, developer experience, real-world usage patterns, registry/discovery gaps, and lifecycle trust signals.

Published Articles

aeshift.com publishes daily commentary on the agent ecosystem, generated through our automated pipeline and reviewed before publication. 13 articles published to date, covering topics from agent sandbox escapes to behavioral evaluation methodology. Additional articles on dacharycarey.com cover:

• Agent-friendly documentation: what it means and why it matters
• The distinction between LLM training data and agent consumption
• Practical implementation guides for documentation teams and software development organizations

Automated Research Infrastructure

Our daily intelligence pipeline is a key differentiator. It runs autonomously across four stages on self-hosted infrastructure:

1. news-gather (06:00 UTC) — Scans RSS feeds, arXiv preprints, and GitHub releases for agent ecosystem developments. Deduplicates against MongoDB Atlas and uses Claude Sonnet to tag and route items to downstream pipelines.
2. research-sourcing (06:20 UTC) — Evaluates items for research potential, identifies connections via Voyage AI vector search, and tracks theme evolution across lifecycle stages (emerging, active, mature, fading, dormant).
3. shift-sourcing (06:40 UTC) — Selects top items for commentary, searches existing coverage, generates parallel drafts with Claude Opus and GPT-5.2, then fact-checks and edits for style. Opens a PR on the aeshift repo with publication-ready drafts.
4. dailies dashboard (07:20 UTC) — Synthesizes pipeline outputs, API usage, research themes, and task status into an operational dashboard for the team.

This infrastructure means we catch developments as they happen, not weeks later. The pipeline has produced 13 published articles on aeshift.com and continuously tracks research themes to inform priorities and upcoming publications.

Research Roadmap

Our current work in documentation and tool quality is the foundation. As the program grows, we plan to expand along four directions:

• Broader ecosystem reports. The agent ecosystem extends well beyond documentation and skills. We’re laying groundwork for reports covering code generation quality, agent platform comparisons, tooling developer experience, and agent performance benchmarks.
• Additional specifications. The Agent-Friendly Documentation Spec addresses one surface where agents interact with the web. We see similar opportunities for agent-friendly standards in other areas: web content more broadly, tool definition formats, and output quality guidelines.
• Registry and scoring infrastructure. The Agent Tool Registry is a first step. We’re working toward infrastructure that provides quality signals, interoperability metadata, and trust indicators across agent tools, servers, and integrations.
• Practitioner education. The research produces findings; the next step is making those findings actionable for documentation teams, platform builders, and developers working with agents day to day.