Your LLMs need a
crossing guard.

Monitor and control native sessions and tool calls in Claude Code, Codex and OpenCode, with deterministic policies or AI agents.

Public release in preparation.

No public install or download is available from this page. This is a website design preview, not a product release. Examples, support labels and metrics illustrate the intended design.

Works with the agents you already use — and says exactly how well

CC Claude Code Proven
CX Codex Proven
OC OpenCode Proven
CP Copilot CLI Registered
GM Gemini CLI Registered

Same binary, same rule file, demonstrated live on three agents across three model families. The two "registered" agents have their hook seams mapped; coverage is probed before we ever call it proven. We never claim a wall we didn't build — that honesty is the product.

What it does

01

One memory

Your project knowledge follows you between agents — files-first, no vendor store. Hit a limit on one, pick up on another with the context intact.

crossing memory — carry
crossing carry --from claude --to codex reading 3 dossiers, 1 open session… ✓ memory injected into codex session ✓ 41 min of context preserved nothing left the machine.
crossing coverage — the honest report
crossing coverage protect-shared-branch rule → per-agent enforcement: claude-code enforced (fail-closed) codex enforced (fail-closed) opencode best-effort (fails open on timeout) copilot-cli unsupported — seam not yet probed

02

One rulebook

Write "never push to main" once. It compiles to each agent's own hook seam — and the coverage report tells you, per rule, per agent, what's truly enforced versus best-effort. Nobody else labels the seams honestly.

03

Every agent, one file

The same rules.json drove a live block on Claude Code, Codex, and OpenCode — three integration styles, one source of truth. Add a runtime by adding a data row, not a rewrite.

rules.json — authored once
{ "id": "protect-shared-branch", "match": { "capability": "git.push_protected" }, "action": "ask", "backstop": "remote-branch-protection"}

How it works

1

Install & attach

One command, then crossing attach wires the hooks into each agent you use.

2

Write one rule

Pick from the library or write a capability + action. It compiles to every agent's seam automatically.

3

See the coverage

The report shows what's enforced, best-effort, or unsupported — per agent. No surprises, no bluffing.

3 agents, proven live
Claude Code · Codex · OpenCode
1 rule file
3 model families, 3 integration styles
0 bytes to a vendor
local-first, non-logging fallback
Read the probes →

No user-count to quote yet — so here's proof of work instead of proof of a crowd. Every claim above links to a probe you can run.

Never get stuck

Your usage limit resets in five hours.
Your deadline doesn't.

When one agent hits the wall mid-task, the paddle goes up — and your session, memory, and rules cross to another agent or a local model. Keep working.

See the limits board →

When the meter runs out

At limit Claude Max · resets 4h 51m

Cross to Codex → session + memory + rules travel with you.

For the team lead who has to answer for it

Honest governance, not a checkbox

Author once, enforce everywhere

Org policy is a signed, versioned bundle — org > repo > user, tighten-only. Compiles locally; the server just distributes it.

Coverage you can audit

Every rule renders its real enforcement per agent. A security review reads the report, not the marketing.

Run it yourself

Self-host the whole thing, with any future managed service subject to a separate announcement. License and contribution terms are being finalized.

We didn't put Open in the name. We put it in the license.

The Comparison Lab

Choosing between agents? We ran the same task on each.

Reproducible head-to-heads — same repo, same rules, same session-carry test — with probe artifacts attached. Updated when the probes re-run.

MatchupStatus
Codex vs Claude CodeProbe-backed
OpenCode vs Claude CodeProbe-backed
Gemini CLI vs Claude CodeRegistered

See all comparisons →

Put a guard at the crossing.

Open-source release in preparation. Explore the design and proposed workflows.