René's URL Explorer Experiment

Title: Security model | Faramesh Docs

Open Graph Title: Security model

X Title: Security model

Description: Threats Faramesh defends against, the mitigations it ships, and explicit limits.

Open Graph Description: Threats Faramesh defends against, the mitigations it ships, and explicit limits.

X Description: Threats Faramesh defends against, the mitigations it ships, and explicit limits.

Keywords:

Faramesh
AI agents
agent governance
governance as code
policy engine
MCP proxy
LangGraph
FPL
decision provenance

Opengraph URL: https://docs.faramesh.dev/security

direct link

Domain: docs.faramesh.dev

application-name	Faramesh
author	Faramesh
creator	Faramesh
publisher	Faramesh
googlebot	index, follow, max-image-preview:large, max-snippet:-1
category	technology
ai-content-declaration	human-authored-technical-documentation
og:site_name	Faramesh Docs
og:locale	en_US
og:image	https://docs.faramesh.dev/og/docs/security/image.png
og:image:width	1200
og:image:height	630
og:image:alt	Security model
og:type	article
twitter:card	summary_large_image
twitter:image	https://docs.faramesh.dev/og/docs/security/image.png

Links:

	https://docs.faramesh.dev/
	https://docs.faramesh.dev/
Why Faramesh	https://docs.faramesh.dev/introduction
Quickstart	https://docs.faramesh.dev/quickstart
Use cases	https://docs.faramesh.dev/use-cases
Workflows	https://docs.faramesh.dev/flows
How Faramesh compares	https://docs.faramesh.dev/compare/alternatives
The governance.fms file	https://docs.faramesh.dev/stack
FPL language reference	https://docs.faramesh.dev/fpl
Providers	https://docs.faramesh.dev/providers
Security model	https://docs.faramesh.dev/security
Denial codes	https://docs.faramesh.dev/errors
Troubleshooting	https://docs.faramesh.dev/troubleshooting
Limitations	https://docs.faramesh.dev/limitations
	https://github.com/faramesh/faramesh-core
Trust boundary	https://docs.faramesh.dev/security#trust-boundary
Guarantees	https://docs.faramesh.dev/security#guarantees
The agent cannot modify policy	https://docs.faramesh.dev/security#the-agent-cannot-modify-policy
The agent cannot kill the daemon	https://docs.faramesh.dev/security#the-agent-cannot-kill-the-daemon
The agent never holds long-lived credentials	https://docs.faramesh.dev/security#the-agent-never-holds-long-lived-credentials
providers reference	https://docs.faramesh.dev/providers
Every decision is tamper-evident	https://docs.faramesh.dev/security#every-decision-is-tamper-evident
The agent cannot exfiltrate to arbitrary hosts	https://docs.faramesh.dev/security#the-agent-cannot-exfiltrate-to-arbitrary-hosts
Async tasks can't slip through	https://docs.faramesh.dev/security#async-tasks-cant-slip-through
Config load-once (no hot edit bypass)	https://docs.faramesh.dev/security#config-load-once-no-hot-edit-bypass
Daemon lifecycle and DAEMON_NOT_READY	https://docs.faramesh.dev/security#daemon-lifecycle-and-daemon_not_ready
UID separation and process isolation	https://docs.faramesh.dev/security#uid-separation-and-process-isolation
Seccomp and eBPF for hostile agents	https://docs.faramesh.dev/security#seccomp-and-ebpf-for-hostile-agents
Defense in depth	https://docs.faramesh.dev/security#defense-in-depth
Recommended posture per environment	https://docs.faramesh.dev/security#recommended-posture-per-environment
What's next	https://docs.faramesh.dev/security#whats-next
Enforcement	https://docs.faramesh.dev/concepts/enforcement
Auditing	https://docs.faramesh.dev/concepts/auditing
Credentials	https://docs.faramesh.dev/concepts/credentials
Limitations	https://docs.faramesh.dev/limitations
Denial codes	https://docs.faramesh.dev/errors
BedrockGovern AWS Bedrock action group invocations with the Faramesh HTTP proxy.	https://docs.faramesh.dev/frameworks/bedrock
Denial codesEvery structured error Faramesh returns to an agent, with payloads and resolution hints.	https://docs.faramesh.dev/errors
Trust boundary	https://docs.faramesh.dev/security#trust-boundary
Guarantees	https://docs.faramesh.dev/security#guarantees
The agent cannot modify policy	https://docs.faramesh.dev/security#the-agent-cannot-modify-policy
The agent cannot kill the daemon	https://docs.faramesh.dev/security#the-agent-cannot-kill-the-daemon
The agent never holds long-lived credentials	https://docs.faramesh.dev/security#the-agent-never-holds-long-lived-credentials
Every decision is tamper-evident	https://docs.faramesh.dev/security#every-decision-is-tamper-evident
The agent cannot exfiltrate to arbitrary hosts	https://docs.faramesh.dev/security#the-agent-cannot-exfiltrate-to-arbitrary-hosts
Async tasks can't slip through	https://docs.faramesh.dev/security#async-tasks-cant-slip-through
Config load-once (no hot edit bypass)	https://docs.faramesh.dev/security#config-load-once-no-hot-edit-bypass
Daemon lifecycle and DAEMON_NOT_READY	https://docs.faramesh.dev/security#daemon-lifecycle-and-daemon_not_ready
UID separation and process isolation	https://docs.faramesh.dev/security#uid-separation-and-process-isolation
Seccomp and eBPF for hostile agents	https://docs.faramesh.dev/security#seccomp-and-ebpf-for-hostile-agents
Defense in depth	https://docs.faramesh.dev/security#defense-in-depth
Recommended posture per environment	https://docs.faramesh.dev/security#recommended-posture-per-environment
What's next	https://docs.faramesh.dev/security#whats-next

Viewport: width=device-width, initial-scale=1

Robots: index, follow

URLs of crawlers that visited me.