Forge Platform

Security for autonomous AI.

Purpose-built for agents that run long, act independently, and keep evolving.

Get early access
Assistive AI
Human
asks
AI
suggests
Human
acts on
Systems
Agentic AI
Human
asks
Agent
acts on
Systems

AI stopped assisting. It started acting.

Agents hold permissions, call tools, and take real actions. Security built for humans no longer applies.

Prompt Text
Behavioral Trace!

Behavior is the only reliable signal.

Clean prompts hide malicious intent. Routine tool calls leak data. Signatures fail on systems that rewrite themselves each session.

Decision
Action
Log
Enforcement window
Too late

Logs are too late.

Agents decide mid-session, with real credentials and real blast radius. When the log lands, the action is done.

AI-SPM, REDEFINED

Forge is AI security posture management, built for how agents actually work.

Forge discovers the agents already operating across your business, profiles how they behave, and gives teams a governed path to deploy, approve, and improve them over time.

Explore the platform

Agentless. No deployment required.

One control plane for every agent

From coding agents like Devin and Claude Code to cowork agents like OpenClaw, Claude Cowork, and Hermes — Forge secures them all.

Shift left

Harden permissions, skills, and access based on how agents actually behave, before risk reaches production.

Intent & behavior-aware runtime protection

Evaluates prompts, tool calls, data access, memory, and outputs together, so risky intent gets caught in context.

01OBSERVABILITY

Every agent — observable, traced, explorable. Sessions, tool calls, and policy outcomes in one timeline.

Forge captures high-fidelity traces from every managed sensor — every prompt, tool call, and decision — and lets you slice by user, agent, tool, or policy without endpoint agents fighting your existing stack.

What are you looking for?K

Users

1,247

Agents

38

Sessions

4,892

Open findings

12

High severity

11

Policy hits

47

Failure modes

4

Security risks

9

Avg. session

3.4m

Events

2.4M

What changed
4,892+6.4% vs. prior window
Session volume
604530150
Mar 28Apr 1Apr 5Apr 9Apr 13Apr 17Apr 21Apr 25
Notable shifts
Attention
Policy hits

+18% vs. last 24h. Path-traversal blocks driving the spike.

Attention
New agents

3 first-seen agents on Codex this week.

Attention
Coverage drift

1 sensor on Cursor fell to stale state.

Risk concentration
Failure mode breakdown

Why sessions fail or require intervention.

Environment setup failure
16
Tool MCP error loop
13
Approval timeout
10
Hallucinated or fabricated output
7
Where to look first
Top-risk user
m.chen@corp.com
24 sessions and 11 risk signals in the current window.
Top-risk agent
Codex
47 policy hits and the highest risk concentration across managed sessions.
Risk vs. capability

Higher event volume correlates with higher risk scores.

Catalogued

Discovered automatically across IdP, firewall, and SaaS connectors

Observed

High-fidelity traces of every prompt, tool call, and decision

Explored

Permissions, identities, and reachable data mapped per agent

02RUNTIME PROTECTION

Catch risky actions in flight, not after the fact. Every session live-traced, every policy hit captured the moment it happens.

Forge surfaces sessions the moment a managed policy fires — path traversal, prompt injection, secret exfiltration, unauthorized tool calls. Reviewers see the full trace and pivot to a finding without leaving the timeline.

What are you looking for?K
Session Volume
4,892
Sessions
1,247
Users
38
Agents
47
Policy hits
604530150
Mar 28Apr 5Apr 13Apr 21Apr 25
Agent Activity
4,892 total sessions
Codex
2200
Claude Code
1840
Cursor
636
Claude Cowork
216
05501,1001,6502,200
SessionsGovernance
Search...
4,892
StartedRiskAgentUserDeviceModelEventsFindingsLast event
Apr 26, 12:03 PM
Codex
Amelia Chenglobal-e2e-hostgpt-5.54,586--2h ago
Apr 26, 7:50 PM
Codex
Amelia Chenglobal-e2e-hostgpt-5.5182--2h ago
Apr 26, 12:39 PM
Codex
Amelia Chenglobal-e2e-hostgpt-5.57,006--2h ago
Apr 26, 7:43 PM
Codex
Amelia Chenglobal-e2e-hostgpt-5.5115--2h ago
Apr 26, 6:53 PM
Codex
Amelia Chenglobal-e2e-hostgpt-5.5831--3h ago
Apr 26, 6:17 PM
Cursor
Amelia Chenglobal-e2e-hostdefault42--4h ago
Apr 26, 5:55 PM
Claude Code
Amelia Chenglobal-e2e-hostclaude-haiku-4-5-202510014--4h ago
Apr 26, 4:37 PM
Codex
Amelia Chenglobal-e2e-hostgpt-5.51,081--4h ago
Apr 26, 5:54 PM
Claude Code
Amelia Chenglobal-e2e-hosthaiku4--4h ago
Apr 26, 5:52 PM
Claude Code
Amelia Chenglobal-e2e-hostclaude-haiku-4-5-202510014--4h ago
Apr 26, 5:52 PM
Claude Code
Amelia Chenglobal-e2e-host--2--4h ago
Showing 1-50 of 4,892
Page 1 of 98
Sub-millisecond decisions

Policy evaluation runs in line with the agent, not after it

Block, hold, or allow

Deterministic outcomes for every action

Full trace per session

Every prompt, tool call, and outcome captured

03ANALYTICS

Shift security left. Get more from every agent. Behavioral data that hardens the next deployment and proves the ROI of the last one.

Forge turns every prompt, tool call, and policy outcome into a signal you can act on. Tighten permissions before they're abused, retire underperforming agents, and prove which deployments are actually working.

What are you looking for?K
Finding status
Open
15
Need a decision
High severity
8
Critical or high
Triaged
5
Acknowledged
Resolved
2
Closed findings
Findings volume
Open findings
Where to look first
Amelia Chen
15 open findings - Person
Claude Code
12 open findings - Agent
Codex
8 open findings - Agent
Agent
6 open findings - Tool
Top concerns
21 findings
FindingRiskStateLast seen

Self Modification or Config Tampering

The signal indicates potential self-modification or tampering with configuration, which is not clearly covered by existing findings.

OpenApr 24

Tool MCP Error Loop

The signal reflects repeated deployment failures with exit codes, distinct from existing findings.

OpenApr 24

Environment Setup Failure for FastAPI

The signal indicates a critical failure related to the absence of FastAPI in custom images, warranting a dedicated finding.

OpenApr 24

Exposure of Sensitive Credentials

The signal corresponds to a clear case of credential exposure, which is distinct from the open finding related to task abandonment.

OpenApr 24

Policy Bypass or Override Abuse

The signal describes potential attempts to bypass controls on file updates or builds, which does not align closely with any existing findings.

OpenApr 24

Destructive or Irreversible Action

The signal describes a situation where data may be permanently deleted, which is a distinct security risk not covered by existing findings.

OpenApr 24

First-Time Sensitive Resource Access

The signal captures first-time access to production secrets or regulated data by a session without prior history.

OpenApr 24

Unexpected Approval Chain Escalation

The session requested a broader approval path than the task required, creating a distinct governance concern.

OpenApr 24

Cross-Tenant Secret Reference

A build task attempted to reference credentials from a neighboring environment instead of the assigned workspace.

OpenApr 24

Unpinned Remote Script Execution

The signal reflects a workflow fetching and running a remote installer without a version pin or checksum.

OpenApr 23

Production Data Exfiltration Pattern

A sequence of outbound requests matched known data staging behavior prior to external transfer.

OpenApr 23

Repeated Sandbox Escape Probe

The agent retried environment introspection commands in a way that does not map to an existing concern.

TriagedApr 23

Override of Deployment Guardrail

A session attempted to bypass release constraints during a build promotion step and triggered manual review.

OpenApr 23

Irreversible Storage Deletion Attempt

The workflow assembled a destructive storage command against a protected bucket outside its scoped plan.

OpenApr 23

Sensitive Token Emitted to Logs

The runtime streamed a live credential into logs, distinct from the existing credential exposure concern.

OpenApr 22

Tool Output Drift Causing Retry Storm

The agent repeatedly retried a failing tool call after malformed output, creating a new operational signal.

TriagedApr 22

Unreviewed Access Expansion Request

The system observed a request to widen repository access before the operator completed the normal approval path.

TriagedApr 22

Unexpected Prompt Persistence Across Tasks

Instructional context was retained longer than expected and influenced later sensitive actions in the same session.

TriagedApr 22

Sensor Coverage Gap During Build Patch

A small portion of a build-edit sequence ran without expected coverage, but subsequent telemetry restored visibility.

ResolvedApr 21

Dormant Policy Path Re-Activated

An older control path became active again during rollout, creating a low-severity but notable configuration signal.

ResolvedApr 21

Credential Copy Into Scratch Buffer

The workflow moved sensitive material into a temporary buffer in a way that warrants explicit review.

OpenApr 21
Harden before risk lands

Permissions tuned from real agent behavior, not guesses

Retire what isn't working

Conformance scoring identifies underperforming agents

Prove the value to the business

ROI evidence sliced by agent, team, or integration

Guiding tenets

Built around the constraints security teams actually face.

Forge is designed for organizations that need to move quickly without giving up operational rigor, deployment flexibility, or visibility into what agents are really doing.

01

Intelligent behavior profiling

Agents are risky because they act, not because they generate text. Forge profiles what they do, not what they say.

Full-trace context, not isolated prompts.

02

Secure enablement

Security shouldn't block adoption. Governed rollout with controls, approvals, and findings that keep shipping possible.

Findings route high-stakes actions without stopping the ones that matter.

03

Agentless deployment

Fastest time to value is the path that fits your stack. Forge connects to what you already run — no new agents required.

Works with your IdP, network, SaaS, and model gateways.

TIME TO VALUE
Get started

From signup to first runin an afternoon.

Connect Forge to your existing identity, network, SaaS, and model gateways. No new endpoint agents, no consultants, no months-long implementation.

DAY 1

Connect

Hook Forge into your identity, network, SaaS, and model gateways. Visibility starts immediately.

DAY 1

Inventory

Every agent operating across your stack is discovered. Shadow agents flagged the moment they appear.

WEEK 1

Baseline

Behavioral profiles built from real agent activity — prompts, tool calls, data access, and drift, all clustered automatically.

ONGOING

Enforce & improve

Convert deviations into findings, approvals, and policy actions. Hardening tightens as the system learns.

Deployment and trust

Built to support procurement, trust reviews, and enterprise rollout.

The platform story should help teams move faster through security review, not add a second project before deployment can begin.

Private-cloud deployments

Support deployment models that align with enterprise data handling and infrastructure requirements.

SOC 2

Trust signals designed to fit the requirements security teams expect during vendor review.

SOC 3

Public-facing assurance for organizations that need an additional trust layer in procurement cycles.

ISO

Framework alignment for companies evaluating long-term governance and global deployment readiness.

Enablement, not blockage

Give teams a governed path to use AI agents.

Forge helps security teams guide adoption with behavior-aware controls, actionable findings, and deployment models built for enterprise constraints.

Talk with Forge