How-to guides
Step-by-step guides for governing AI agents with AgentGarm: policy rollout, approvals, secrets, cost and recovery; CI and Central Lite fleets in preview.
The guides are written for macOS and Linux. Windows is a preview, and no guide walks it yet.
61 guides
Getting started
For one machine: bring your agents under AgentGarm, with hooks or without, start a policy in audit mode, check that protection is on, turn on the macOS sentinels, follow what it denies or asks, run it on Windows, see the local AI runtimes on it and keep it up to date. 6 of its 12 guides are in preview.
Guide 1
Watch your agent for a week, then enforce
Start a policy in audit mode, see what enforcing it would have blocked in your own history, fix the rules that get in the way, then switch it to enforce.
Guide 2
Govern Claude Code
Install the Claude Code pack, put AgentGarm's hooks in Claude Code's managed settings, run it through the launcher and proxy, and check what is enforced.
Guide 3
Govern Codex CLI
Install the Codex pack, add its hooks to Codex's managed requirements.toml on a root install, start Codex through the launcher, and check what is enforced.
Guide 4
Govern Cursor, Copilot CLI, Devin, Cline and otherspreview
Install your agent's policy pack in audit mode, put AgentGarm's hooks where that agent reads them, and read what its hooks cannot do.
Guide 5
Govern an agent with no hooks
Run an agent AgentGarm has no hook adapter for inside the same sandbox and proxy, and know which of your policy's rules that leaves unenforced.
Guide 6
Route other model providers through the proxypreview
Send Gemini, Bedrock, Vertex, Azure, OpenRouter, DeepSeek and local-runtime calls through the proxy, accounted, budgeted and checked for secrets.
Guide 7
Check that protection is on
See whether AgentGarm's services are up, what assurance level each agent session has and why, and what to do about a bypass alert.
Guide 8
Turn on the macOS sentinelspreview
Activate AgentGarm's Endpoint Security and Network Extension on a Mac, approve them by MDM or by hand, and check that they hold the agent's processes.
Guide 9
Get told when an agent is denied or askspreview
Follow each denial, approval ask and alert as it happens: in a terminal, as desktop notifications, or through your own Linux notifier.
Guide 10
Run AgentGarm on Windowspreview
Install the Windows preview, start agents in its Job Object sandbox, use the tray, know what Windows enforces next to macOS and Linux, and remove it.
Guide 11
See the AI runtimes and desktop AI apps on a machinepreview
List the local model runtimes and desktop AI apps on a machine, see which ones no governed session used, and route local model calls through the proxy.
Guide 12
Check for and verify AgentGarm updates
Find the newest AgentGarm release, check its signature and checksums before you run it, never go back to an older one, and upgrade in place.
Writing policy
For whoever writes a machine's policy: which commands, files and hosts an agent may use, Touch ID and Windows approvals, poisoned configs, agent versions, and testing a policy before it ships. Of its 7 guides, 1 is in preview and 4 have preview steps.
Guide 13
Allow, ask or deny shell commands
Write command rules on the parsed command and its wrappers, not its text, choose between asking and denying, and replay commands before you rely on them.
Guide 14
Limit what an agent can read, write and reachpreview steps
Decide which files an agent may read and write and which hosts it may reach, know which layer holds each rule, and recognise a denial.
Guide 15
Test a policy before you ship it
Write what a policy must allow, ask about and deny as a file of events, replay it before you install the policy, and keep both in your repository.
Guide 16
Approve risky actions with Touch IDpreview steps
Enrol a Touch ID key on your Mac and approve each command your policy asks about with your fingerprint, for that one call only.
Guide 17
Approve risky actions on Windowspreview
Enrol a security key on a Windows PC and approve each command your policy asks about through Windows' own WebAuthn; Windows Hello itself comes later.
Guide 18
Catch poisoned agent configs before launchpreview steps
Scan the instruction, settings, hook, plugin and MCP files your agents read for hidden text and auto-approval switches, and gate CI on the result.
Guide 19
Keep agent versions in checkpreview steps
List your installed agents against the versions AgentGarm verified their hooks for, and bring an agent that updated past that range back in range.
Protecting secrets and data
Keep secrets, tokens and your browser profile from the agent and its model provider: prompt redaction, what agents upload to paste sites and webhooks, sealed cloud and MCP credentials, the MCP gateway, prompt injection in MCP tool output, browser agents, computer-use agents and what agents can capture on screen; gateway keys and the signed MCP registry need Central Lite. Of its 11 guides, 10 are in preview and 1 has preview steps.
Guide 20
Keep secrets and card numbers out of promptspreview steps
See which credentials your agent sends its model provider, have the proxy replace them with cache-safe placeholders, and deny tool calls that carry one.
Guide 21
Check what agents upload to paste sites and webhookspreview
Have AgentGarm read what launched agents send to paste sites, file sharing and webhooks, record or stop secrets, and fix tools that refuse its certificate.
Guide 22
Give Claude Code a per-developer gateway keypreview
Give each developer their own gateway key from Central Lite and let Claude Code fetch it through apiKeyHelper; a claude.ai login is never routed.
Guide 23
Give agents short-lived AWS, GCP and SSH credentialspreview
Seal AWS, GCP and SSH credentials in agentgarmd; each task started with agentgarm run gets short-lived ones, and the sealed ones never reach the agent.
Guide 24
Run MCP servers through the gatewaypreview
List the MCP servers an agent may use, pinned to a package and hash, and run each through the gateway, where agentgarmd decides every tool call first.
Guide 25
Sign in to MCP servers without exposing tokenspreview
Give your MCP servers their API keys and OAuth sign-ins from your own terminal; agentgarmd seals each one and the agent never sees a token.
Guide 26
Allow only vetted MCP serverspreview
Mirror the MCP registry, have Central Lite sign it for your machines, and let agents launch only servers resolved from that signed snapshot.
Guide 27
Contain prompt injection in MCP tool outputpreview
Strip hidden text from MCP tool results, hold back poisoned tool descriptions, pin tool definitions, and limit what a hijacked agent can do next.
Guide 28
Govern browser agentspreview
Put AgentGarm's browser shim in front of the browsers your agent launches, so each launch gets a fresh profile and its downloads wait in quarantine.
Guide 29
Govern computer-use agentspreview
Let an agent drive your Mac's mouse and keyboard while AgentGarm's airbag drops its input on password managers, settings and other protected targets.
Guide 30
Watch what agents can capturepreview
See which agents can capture your Mac's screen, and alert, pause or end a task when a password manager or other sensitive app comes to the front.
Cost
For one machine: budget each task, see what each task cost by provider and model, and reconcile the spend with your provider's own reports. Of its 3 guides, 1 is in preview and 1 has preview steps.
Guide 31
Set spend budgets and stop runaway loops
Give each agent task a cost budget and a token rate, see what the agent meets at the cap, and stop loops and runaway subagents with limits of their own.
Guide 32
See what each task costpreview steps
Route an agent's model calls through AgentGarm so each one is counted and priced, and read the spend by provider and model.
Guide 33
Reconcile spend with provider invoicespreview
Reconcile a day of this machine's model calls against your provider's own usage and cost reports, read where they differ, and export FOCUS rows.
Investigating and recovering
On one machine, when you need to know what happened: trace what a task did, undo its changes, find out why an agent stopped, verify and export the audit log, export a task's evidence, send events to your SIEM from each machine or from Central Lite, and write a support bundle. Of its 8 guides, 4 are in preview and 3 have preview steps.
Guide 34
Undo what an agent changedpreview steps
Preview what rolling back an agent's task would restore and keep, then apply the rollback from the checkpoint taken before the agent started.
Guide 35
Find out why your agent stoppedpreview
Tell whether your agent was suspended, killed, locked out by a kill order or stopped by its budget, get your work back, and find out who can lift it.
Guide 36
Trace what one agent task didpreview steps
Find an agent's task in the dashboard, read every decision its policy made with the rule behind it, and follow the run from first decision to last.
Guide 37
Export and verify the audit log
Export AgentGarm's hash-chained audit log, check it against the node key that signs its heads, and know what a passing or broken check tells you.
Guide 38
Export evidence for one taskpreview
Write one task's signed evidence bundle on its machine, verify it with the node's key, hold its records, and hand it to someone else to check.
Guide 39
Send agent events to your SIEMpreview
Turn on agentgarmd's exporters: OCSF events to your SIEM, OTel GenAI spans and token metrics to your collector, and notice when an export stops.
Guide 40
Send the fleet's security events to Splunk or Sentinelpreview
Have Central Lite forward the fleet's security and audit events, its own audit log and its alerts as OCSF to Splunk or Microsoft Sentinel.
Guide 41
Send a redacted support bundlepreview steps
Write a JSON file of diagnostics with your home directory, user and host names and the secrets AgentGarm recognises redacted, sent only when you share it.
CI, containers and servers
For machines with no one at the keyboard: a server or container whose asks are denied instead of waiting, and a CI job whose risky step you approve from your laptop. Of its 2 guides, 1 is in preview and 1 has preview steps.
Guide 42
Govern agents on a headless server or containerpreview steps
Check what a server or container offers, set a policy to the headless profile so every ask is denied instead of waiting, and test it before you install it.
Guide 43
Approve a CI job's risky step from your laptoppreview
Mark the commands a CI job may wait on a person for, and approve the job's request from the console with your passkey, for exactly that call, once.
Running a fleet with Central Lite
For a Central Lite admin: tour its console in the demo, stand it up, keep its keys in an HSM, deploy machines with MDM, sign in through your identity provider, provision people with SCIM, give roles, enrol machines, roll out policy and outbound inspection, stop one task or agents across the fleet, investigate a task's incident, attribute spend, and manage identities, retention, works-council packs and backups. Every guide here is in preview.
Guide 44
Tour the fleet console in the demopreview
Walk Central Lite's console on a fictional company's synthetic data: its machines, an incident, a kill order, costs, outbound inspection and access.
Guide 45
Stand up Central Litepreview
Set up Central Lite on one Linux host: its org and keys, its HTTPS port, a first admin and a backup passkey, and its first encrypted backup.
Guide 46
Keep Central Lite's keys in a hardware security modulepreview
Create Central Lite's org with every key wrapped under a key that never leaves your PKCS#11 token, and keep what a restore needs backed up.
Guide 47
Sign in to the console with your identity providerpreview
Let admins open Central Lite's console through Okta, Entra ID or another OIDC provider, with roles from its groups and a passkey still behind every change.
Guide 48
Provision and remove people with SCIMpreview
Let Okta or Entra ID create the people Central Lite knows, give them console roles from their groups, and end their access when they leave.
Guide 49
Give people the right access in the consolepreview
Give each Central Lite admin one of six roles, for the whole org or a host group, and know who may request a kill order and who holds each approver key.
Guide 50
Enrol developer machinespreview
Enrol a machine in Central Lite with a one-time token that pins Central Lite's key, see it report in the inventory, and let its users enrol presence keys.
Guide 51
Deploy AgentGarm with MDMpreview
Roll AgentGarm out to managed Macs, Linux machines and Windows PCs: the packages, the services, the macOS approvals and the browser extension.
Guide 52
Roll out policy from git with canariespreview
Keep agents.d policies in git and let Central Lite roll each signed commit out to canaries, then the fleet, halting by itself when machines report trouble.
Guide 53
Roll out outbound inspection across a fleetpreview
List the sites your fleet's agents should not send secrets to, run them in Audit, read what Block would stop, then block through a canary rollout.
Guide 54
Stop an agent task on any machinepreview
Kill or suspend one agent task from the Central Lite host, check that the node applied it, and revoke a machine you no longer trust.
Guide 55
Stop agents across a fleetpreview
Suspend, terminate or lock down agents on one machine, one agent, one person or the whole fleet, with an order two approver keys sign and each node checks.
Guide 56
Investigate a task across the fleetpreview
Read a task's timeline in Central Lite with its gaps, export a bundle the org root's evidence key signs, verify it offline and hold its records.
Guide 57
Give every agent an owner and an identitypreview
Register each person and agent in Central Lite's identity registry, issue and revoke gateway keys, and see which identity every task belongs to.
Guide 58
Attribute AI spend by machine and personpreview
Read your enrolled machines' model spend by day, machine, provider and model, and reconcile it against your providers' own usage and cost reports.
Guide 59
Set retention, legal holds and data-subject requestspreview
Set how long Central Lite keeps each class of record, put a legal hold on the records a case needs, and answer a person's access or erasure request.
Guide 60
Prepare a works-council or DPIA packpreview
Generate templates that describe what your deployment collects, keeps and sends, and switch the fleet to the non-monitoring profile during talks.
Guide 61
Back up and restore Central Litepreview
Write Central Lite's encrypted backups by hand and on a schedule, and restore one so that no enrolled node refuses what Central Lite signs next.