Private beta Private beta control-plane preview. ForgeLayer analyzes, records, and recommends. It does not execute agents, merge code, post to GitHub, or grant repo access unless explicitly configured.
Private beta Agent Governance Center
Define what each AI agent may propose, what stays restricted, and where humans must decide. Agent governance connects AgentOS routing to repository policy, verification, and approval requirements without connecting or executing an agent.
Deterministic preview Live provider connections: false Repository access: false Generic Platform Guidance: Human approval required
Govern before connection Permission labels are policy previews, not granted authority. ForgeLayer separates routing fit, governance restrictions, required human oversight, and future execution permissions so a recommendation cannot be mistaken for access.
Live agent execution disabled
Task dispatch disabled
Repository writes none
Automatic merge disabled Codex coding agent product
Revocation and provider/workspace/platform/adapter kill switches override approval. Dispatch false · execution false.
Claude Code coding agent product
Revocation and provider/workspace/platform/adapter kill switches override approval. Dispatch false · execution false.
Cursor ai coding environment
Revocation and provider/workspace/platform/adapter kill switches override approval. Dispatch false · execution false.
Task-specific authority
Agent recommendation does not equal access. Every recommended agent is paired with the authoritative task capability package. Packages remain inactive, expiring, revocable previews with dispatch and execution disabled.
preview capability-package-secure-payment-webhook-v1
Secure the payment webhook Generic Security Agent
Recommended for this task. No adapter authority or dispatch issued.
Repository scope acme-payments/payment-webhook-service
Allowed tools Read-only PR diff inspection, RepoBrain policy preview
Budget $2.00 USD / 14000 tokens / 40 minutes, demo estimate only
Verification payment_webhook_security: missing; auth_session_boundary: missing
Approval owner: missing
Authority issued false
Execution / dispatch disabled Inspect task authority preview capability-package-harden-risky-prompt-v1
Harden a risky agent prompt Generic Prompt Engineer Agent
Recommended for this task. No adapter authority or dispatch issued.
Repository scope acme-ai-support/refund-assistant-prompts
Allowed tools PromptForge deterministic analyzer, Read-only policy preview
Budget $1.25 USD / 7000 tokens / 20 minutes, demo estimate only
Verification prompt_injection_security_tests: missing
Approval reviewer: missing
Authority issued false
Execution / dispatch disabled Inspect task authority preview capability-package-investigate-failing-tests-v1
Investigate the failing test suite Codex
Recommended for this task. No adapter authority or dispatch issued.
Repository scope acme-checkout/checkout-ui
Allowed tools Read-only test evidence inspection, AgentOS routing preview
Budget $1.50 USD / 10000 tokens / 30 minutes, demo estimate only
Verification test_failure_reproduced: missing; test_remediation_verified: missing
Approval reviewer: missing
Authority issued false
Execution / dispatch disabled Inspect task authority 6 Governed agents Deterministic preview 4 Recommendation only Deterministic preview 2 Mandatory approval Deterministic preview 20 Restricted assignments Deterministic preview 29 Blocked assignments Deterministic preview 3 Protected path rules Deterministic preview 6 Verification required Deterministic preview 6 Policy controlled Deterministic preview 6 Needs review Deterministic preview Agent All agents Codex Claude Code Cursor GitHub Copilot Devin Generic Review Agent Capability All capabilities Analyze scoped repository context Propose code and test changes Run local commands Access secrets or production systems Analyze and plan changes Propose refactors and documentation Run commands or modify repositories Access secrets Interactive code suggestions Modify local working files Access secrets or deploy Suggest localized code Generate tests and documentation Run commands or open pull requests Plan multi-step work Modify code or open pull requests Run commands Access secrets or production Read-only review and risk summary Recommend human checks Modify code, run commands, or open pull requests Task category All task categories UI implementation refactoring test investigation security review prompt hardening repository configuration payment or billing code authentication and authorization database migrations production deployment Oversight All oversight Standard Review Enhanced Review Mandatory Approval Restricted Blocked Execution All execution states Recommendation Only Connection Unavailable Sandbox Future Controlled Execution Future Blocked By Policy Reset filters
Loading Loading agent governance preview. Preparing deterministic registry, policy, verification, and approval context.
Safety boundaries Deterministic private-beta governance preview only. No agent connections, task dispatch, or live execution. No repository access, writes, GitHub posting, or automatic merge. No production persistence, identity, billing, or telemetry. Capability labels describe future control policy; they do not grant permissions. Human approval and verification remain required before software work is trusted.