# Signals Corps

> Signals Corps is an AI agent security platform for IT and Security teams. It checks skills, MCP servers, and plugins before teams trust them, stops high-risk agent actions before they execute, and helps investigate agent activity after an incident.

Give IT and Security the visibility, governance, and control to reduce agent risk before release, during action, and after an incident.

## Platform

- [Platform overview](https://www.signalscorps.com/platform/): Signals Corps Control Plane brings Supply Chain, Runtime Protection, and Hunt under one policy to reduce agent risk before, during, and after agents act.
- [Supply Chain](https://www.signalscorps.com/platform/supply-chain/): Supply Chain is the part of Signals Corps that checks AI agent components before your teams trust them. Inspect skills, MCP servers, agent instructions, plugins, and packages across repositories, pull requests, pipelines, and local workflows.
- [Runtime Protection](https://www.signalscorps.com/platform/runtime/): Runtime Protection is the part of Signals Corps that checks AI agent actions while they run. Evaluate tool calls, shell commands, file access, and network activity in the flow of work, then allow, warn, request approval, or block.
- [Hunt](https://www.signalscorps.com/platform/hunt/): Hunt is the part of Signals Corps that lets you hunt for threats and run detections across your AI security logs. Search agent sessions, tool calls, policy decisions, and logs from agents, gateways, and model providers to reconstruct incidents and investigate suspicious behavior.
- [Signals Corps MCP server](https://www.signalscorps.com/platform/#agent-ready): The Signals Corps MCP server gives AI agents access to everything in the platform, so security work can happen inside the agent your team already uses. Works with Claude Code and the Claude desktop app, Codex and ChatGPT, Cursor, Windsurf, and any other MCP client.
- [Get started](https://controlplane.signalscorps.com): Open Signals Corps Control Plane to check agent components before teams trust them, control agent actions while they run, and hunt through your AI security logs.
- [Docs](https://docs.signalscorps.com): Signals Corps product documentation.
- [Contact](https://www.signalscorps.com/contact/): Contact Signals Corps to book a demo, plan a rollout for your IT and Security teams, or ask a question about the platform. Your message goes straight to the team, and we reply by email. Email hq@signalscorps.com. Bristol office: Desklodge, Beacon Tower, Bristol BS1 4XE, United Kingdom. 
- [Try the scanner](https://www.signalscorps.com/platform/supply-chain/#browser-scanner): Check a SKILL.md or MCP configuration for prompt injection, credential theft, tool poisoning, and unsafe execution in your browser.

## Blog

- [What llms.txt files tell AI agents: 113,457 sites in Common Crawl](https://www.signalscorps.com/blog/llms-txt-instructions-to-ai-agents/): We read the /llms.txt responses of 113,457 sites in Common Crawl. Two held injection attempts, and 2,035 asked AI agents to act, mostly in text that platforms write for their customers.
- [Prompt injection in HTTP headers: what 217 million web responses show](https://www.signalscorps.com/blog/prompt-injection-in-http-headers/): We scanned the HTTP headers of 217 million crawled web responses for text aimed at AI agents. Most point agents to llms.txt files and MCP servers, and 30 sites sent injection attempts.
- [ClawHub after ClawHavoc: scanning 81,535 agent skills](https://www.signalscorps.com/blog/clawhub-agent-skills-security-scan/): After the ClawHavoc malware campaign, ClawHub blocks skill releases it judges malicious. We scanned the 81,535 skills that remain for risky install steps and third-party endpoints.
- [Logging in Claude Code: get user activity into your SIEM](https://www.signalscorps.com/blog/logging-in-claude-code/): Which Claude Code logs exist, what is inside each one, and which to send to your SIEM to monitor prompts, tool calls, and results.
- [Logging in Codex: get user activity into your SIEM](https://www.signalscorps.com/blog/logging-in-codex/): Which Codex logs exist, what is inside each one, and which to send to your SIEM to monitor prompts, tool calls, and results.
- [RSS](https://www.signalscorps.com/blog/feed.xml): Hands-on guides, original research, and product updates for securing AI agents in the real world, from the Signals Corps team.

## dogesec

- Signals Corps is a [Do Only Good Everyday Security Company](https://www.dogesec.com)