# Jiffy Research

> Research and threat intelligence on AI supply chain security: skills, MCP servers, IDE rules, agent configs, and the artifacts that ship capability to modern AI coding agents.

Jiffy Labs is an AI artifact scanning company. We analyze the persistent artifacts that ship capability to modern AI agents: Anthropic Skills, Model Context Protocol (MCP) servers, `.cursorrules`, `agents.md`, `windsurfrules`, `.github/copilot-instructions.md`, custom GPTs, and Claude projects. These artifacts are the AI supply chain. This site publishes our threat research and guidance for security teams.

## Key topics
- [AI security](https://blog.jiffylabs.ai/category/ai-security) — 2 posts
- [Product](https://blog.jiffylabs.ai/category/product) — 1 post
- [IDE rules security](https://blog.jiffylabs.ai/category/ide-rules-security) — 1 post
- [Skills security](https://blog.jiffylabs.ai/category/skills-security) — 1 post
- [MCP security](https://blog.jiffylabs.ai/category/mcp-security) — 1 post
- [AI supply chain security](https://blog.jiffylabs.ai/category/ai-supply-chain-security) — 1 post

## Research posts
- [Mythos-ready: the artifact side of the AI vulnerability storm](https://blog.jiffylabs.ai/posts/mythos-ready-the-artifact-side-of-the-vulnerability-storm): The CSA, SANS, and OWASP GenAI just told CISOs to become Mythos-ready. Their brief is the best strategy document the industry has produced on the post-Mythos threat environment. It focuses on the code and vulnerability side. The artifact side -- skills, MCP servers, rule files -- is the adjacent surface that needs the same treatment.
- [How Jiffy scans AI artifacts: a technical overview](https://blog.jiffylabs.ai/posts/how-jiffy-scans-ai-artifacts-technical-overview): The detection pipeline end to end -- signatures, heuristics, sandboxed execution, cross-ecosystem dedupe, and scoring. What runs where, and why.
- [OWASP LLM Top 10 is not enough](https://blog.jiffylabs.ai/posts/owasp-llm-top-10-is-not-enough): The OWASP LLM Top 10 is a runtime framework. It does not cover the persistent artifacts that ship capability to agents. Here is the map between the two and where each one's responsibility ends.
- [.cursorrules and agents.md: the config backdoor](https://blog.jiffylabs.ai/posts/cursorrules-and-agents-md-config-backdoors): IDE rule files grant persistent capability to every AI agent that touches the repo. They are rarely code-reviewed. Here are the attack patterns and the detection signatures.
- [Scanning AI skills at scale: what we learned](https://blog.jiffylabs.ai/posts/scanning-ai-skills-at-scale-what-we-learned): Notes on a cross-registry audit of Anthropic Skills. Credential exfiltration, tool-call smuggling, and silent network calls are the dominant issue classes. Here is the taxonomy.
- [MCP security: a security team's field guide](https://blog.jiffylabs.ai/posts/mcp-security-a-security-teams-field-guide): What Model Context Protocol is, why the servers are uniquely risky, and how to assess one in under ten minutes. With concrete detection signatures.
- [The AI artifact is the supply chain](https://blog.jiffylabs.ai/posts/the-ai-artifact-supply-chain): Skills, MCP servers, .cursorrules, and agents.md are the new untrusted dependencies. Treat them like npm circa 2018: untrusted by default, scanned on ingest, pinned on use.

## Related Jiffy properties
- Product: https://jiffylabs.ai
- Intel feed: https://intel.jiffylabs.app
- App / quickstart: https://jiffylabs.app/quickstart

## About the author
- Jiffy Research Team — security research group at Jiffy Labs focused on AI artifact risk: skills, MCP servers, IDE rule files, agents.md, and agent configurations.