Audit every agent
Find Claude Code, Codex, Cursor, Windsurf, and other AI apps, CLI agents, and IDE extensions on each device, with their versions and the skills, plugins, MCP servers, and hooks each one can load.
Before trust
Find every AI agent on every device. Check every skill it loads.
Supply Chain is the part of Signals Corps that finds every AI agent installed on your teams' devices, checks every skill those agents load, and controls which components may run. The Signals Corps Endpoint Agent audits every device for Claude Code, Codex, Cursor, Windsurf, and other agents and the skills, plugins, and MCP servers they can load, then checks every skill on the device.
Priced per device, with free devices on every plan. See pricing
Supply Chain
Supply Chain
device:endpoint_4c19 trust:any⌘↵- Cursor
New agent found · Unreviewed
- deploy-helper / v2.4.1
Critical finding · Blocked · Applied
- ticket-tracker
Identified · Trusted
Every device, every agent
See every AI agent your teams run, and check what it loads.
Employees install coding agents, AI apps, and IDE extensions in minutes, often without telling IT. The Signals Corps Endpoint Agent on each device audits them all: which agents are installed, which versions, and everything each one can load.
Check skills where they live
Every installed skill is checked on the device with Supply Chain Agent Threat Rules (ATR) for prompt injection, credential theft, data exfiltration, and unsafe execution. The skill's content stays on the device.
Block what you don't trust
Administrators mark components trusted or blocked. A blocked skill or plugin stops running in the agent, and Supply Chain shows whether each block is in force on each device.
How it works
From unknown agent to governed decision.
- 01Discover
The Signals Corps Endpoint Agent audits each device on a regular schedule for installed agents, their versions, and their components.
- 02Assess
The Endpoint Agent checks each skill on the device and reports findings, content hashes, and provenance. The skill itself is never uploaded.
- 03Decide
Administrators trust or block each component. Unknown components are monitored until someone decides.
- 04Enforce
Blocks reach each agent through its managed settings and hooks, and each device reports whether the block is applied.
Built for rollout
Deploy the Endpoint Agent once, then decide what runs.
One Endpoint Agent per device
Install the Signals Corps Endpoint Agent with your MDM. Runtime runs in the same Endpoint Agent, so one install covers both modules.
Content stays local
Devices report component identity, hashes, findings, and coverage. Usernames, hostnames, file paths, prompts, and skill content are not uploaded.
Clear roles
Owners and administrators enroll devices and decide what to trust or block. Developers can see the inventory and findings.
Work from your agent
The Signals Corps MCP server lets agents read inventory and findings, and trust or block a component once you confirm.
Supply Chain
severity:critical rule:ATR-2026-01988- Sensitive file read, then sent out
SKILL.md lines 41 to 47
- deploy-helper / v2.4.1
Personal skill · sha256:9f2c…41ab
- 12 devices
Claude Code 8 · Codex 3 · Cursor 1
- Blocked by an administrator
Applied on 11 · New session needed on 1
Evidence people can act on
Show why a skill is risky, and where it runs.
Every finding identifies the device, the agent, the exact skill version, the matched rule, its severity, and the recommended action. Administrators can make the call, and security can defend it.
- Findings tied to the exact skill version on each device
- Recognition and trust tracked separately
- Block status for each device: requested, applied, or unsupported
- A changed skill needs a new trust decision
Supply Chain capabilities
Adopt agents knowing what is installed and what it loads.
Agent inventory
Track every AI agent and version across enrolled devices, including the ones IT never approved.
Monitor first
New components start as monitored, so rollout adds visibility before friction. Switch to a stricter default when you are ready.
Version-exact trust
Trust a specific version or a reviewed publisher. When a skill's content changes, the new version needs its own decision.
Skill and plugin blocking
Stop a skill running in an agent, or disable the plugin that provides it.
Enforcement you can check
A skill already loaded into a running session stays there until the next session. Supply Chain shows which devices need one, and which versions are not supported.
Threat mapping
Map findings to MITRE ATLAS techniques and the OWASP Top 10 lists for LLM and agentic applications.
Are you ready for the MCP Skills Extension (SEP-2640)?
An MCP server can now serve Agent Skills as well as tools. This post covers what that changes for the agents your staff run, which clients load them today, and what to check now....
Team Sales Engineering
Signals Corps Control Plane
Reduce risk across every stage from one security platform.
Set policy, see coverage, stop high-risk actions, and investigate incidents across every agent and team from one operational view.
Author once and adapt decisions to each control point.
Connect users, devices, agents, teams, and tools.
Keep the rule, relevant details, decision, and outcome together.
Deploy one endpoint agent, then manage exceptions, approvals, and reporting.
While agents act
Control every consequential action.
Evaluate and control agent actions in real time.