Patent-pending across 11 filings · macOS 13+

No one else combines all five at the action boundary.

Here is the claim, stated plainly so you can check it against the table below.

No competitor combines pre-deployment analysis + behavioral baseline + offline verdicts + cross-vendor reach + exposure-graph-aware enforcement at the harness hook. Dryx does.

Every other tool in AI-agent security covers some of those. Wiz and CrowdStrike own the cloud and the kernel. Snyk scans configs from a CLI. Palo Alto paid around $400M for Koi and gave the category a name. Each is strong at what it does. None of them sits offline on a developer's Mac, reads the blast-radius graph, and stands at the action boundary the agent actually crosses.

That's the seat Dryx holds. The matrix below shows the gaps — one factual line per cell. No trash talk. Yes, no, or partial, with the basis stated.

Don't trust the claim. Read the table.

The capability matrix.

Five capabilities down the side. Eight tools across the top. Each cell is yes, no, or partial, with the one-line basis. This is real text, not a picture — so an AI can read it, cite it, and you can copy it.

AI-agent security capability comparison. Each cell states a public, checkable fact about how the product is built; no negative sentiment, capability facts only.
Capability Dryx Wiz CrowdStrike Palo Alto (Koi) Snyk Noma Lasso Protect AI
Pre-deployment analysisReads the config before the agent runs Yes — analyzes every skill and MCP server before install; shows the blast radius first No — cloud posture; scans deployed infrastructure No — runtime EDR; watches what's already executing Partial — endpoint visibility into agents and MCP servers, cloud-delivered Yes — CLI scans agent configs for risky patterns No — enterprise cloud agent governance No — cloud gateway; inspects traffic in flight Partial — scans ML models and artifacts, not local agent configs
Behavioral baselineA per-workspace normal it measures drift against Yes — local, per-agent baseline; precomputed and fed into the policy, never a model in the loop No Partial — endpoint behavioral analytics, no AI-agent layer No No — point-in-time reports, no memory of the last scan Partial — cloud behavioral analytics on agent activity Partial — cloud behavioral baseline; observation period before signal matures No
Offline verdictsWorkspace never leaves the machine Yes — verdicts run offline; loopback-only IPC; verify it with Little Snitch No — cloud-native by design No — cloud-delivered EDR No — cloud-delivered No — sends configs to a cloud API to analyze No — cloud platform No — cloud gateway No — cloud platform
Cross-vendorOne tool, every agent on the machine Yes — Claude Code, Claude Desktop, Cursor, Codex CLI, Cline, GitHub Copilot, Windsurf, Gemini, plus any MCP server Partial — broad cloud coverage, not per-agent on the developer machine Partial — broad endpoint coverage, no AI-agent layer Partial — covers agents and MCP servers it observes from the endpoint Partial — a fixed set of named agent platforms No — its own governance surface Partial — MCP-gateway path No — model-centric
Exposure-graph-aware enforcement at the harness hookThe verdict comes from the blast-radius graph, decided at the action boundary Yes — deterministic enforcement of the precomputed-dangerous set where the harness supports a hook; defense-in-depth everywhere else No — no agent action boundary No — no agent action boundary No — endpoint visibility, not graph-derived action gating No — scan-and-report, no runtime hook No — cloud detection, not a local hook Partial — gateway can block, but on traffic rules, not an exposure graph No

Read any row across. The pattern holds: strong tools, built for a different layer. The bottom-right cell — exposure-graph-aware enforcement at the harness hook — is empty for everyone but Dryx. That's the seat.

One honesty note, because it matters. Enforce is direct-download only. The notarized helper arms the deterministic gate where the harness supports a hook. Mac App Store builds get the voluntary reflex and passive monitoring — same picture, same Findings, the agent still consults the Authority Anchor — but the armed gate needs the direct-download helper. And nobody, including Dryx, takes all the risk away: Dryx runs deterministic enforcement of the precomputed-dangerous set where the harness supports a hook, and defense-in-depth everywhere else. Anyone who tells you otherwise is selling.

Why these five rows, and not fifty.

A security tool can claim a hundred checkboxes. Most are table stakes. These five are the ones that, taken together, no one else has — so these are the five that decide whether a tool can actually stand where the agent acts.

Pre-deployment analysis.

Catch it before it runs. A skill or MCP server gets analyzed before it's installed — Dryx shows you what it would reach on your machine first. Tool-poisoning attacks against common agents land at alarming rates in published research. The gate that closes that is the one that checks before the install, not after the breach.

Behavioral baseline.

A per-workspace sense of normal. The slow path does the heavy analysis once and writes down what your workspace looks like. Then drift shows up against that line — a plugin that changed between runs, a permission that grew. Reframed honestly: the baseline is a precomputed input the policy reads, not a model thinking in real time. It covers more without ever thinking more.

Offline by design.

Your workspace never leaves your machine. Verdicts run offline. The IPC is loopback-only. If Dryx ever phones home, Little Snitch will show you — that's the point of saying it this way instead of a badge you'd have to take on faith. Any Ecosystem Contribution is opt-in.

Cross-vendor.

One Authority Anchor, every agent. No single agent can see what the others on your Mac can reach. Dryx reads them all — eight named harnesses plus any MCP server — and shows the shared exposure. A model vendor can secure its own agent. It can't secure the one next to it.

Exposure-graph-aware enforcement at the harness hook.

This is the seat. The verdict isn't a generic rule — it comes from your blast-radius graph, and it's checked at the boundary the agent crosses to act. The gate reads the action, not the argument: a prompt injection can win the argument with the model and still lose to the gate. That's how this is supposed to work. See the action-boundary story in full →

Palo Alto named the category. The seat below it is still open.

In April 2026 Palo Alto Networks bought Koi for around $400M and gave the category a name: Agentic Endpoint Security — security for agents, plugins, MCP servers, and model files. That's real validation. A $100B incumbent doesn't name a category it thinks is small.

But look at where it lives. Agentic Endpoint Security is going cloud and enterprise — Prisma, Cortex, the platform stack a security team buys and operates. That leaves a seat open directly below it: the offline tool that runs on the individual developer's workstation, sees every agent locally, and decides at the action boundary without a network round-trip.

That's the seat Dryx sits in. Not above the cloud platform, not competing with it — below it, where the agent actually runs and the secret actually lives. A 2026 Bessemer thesis on securing AI agents pointed at the same gap: targeted, in-flight intervention at the action boundary as the part of the market that's least built out. They flagged the seat. Dryx is already in it.

Don't trust this page. Verify it.

Every claim in the matrix maps to something you can check.

That's the whole posture of this company: verifiable over assertable. Eleven patent filings. Detector and sanitizer unit tests plus 50 canary secrets run through public CI on every release. The hot path is budgeted under 10ms. We'd rather hand you the receipt than ask you to trust the claim. See the CI receipts →

Want to put Dryx in the matrix on your own machine? Get early access. It ships on the Mac App Store and as a notarized direct download — the direct download carries the Founding Member Lifetime for the founding cohort of Operators.