Can OrchestrAI read secrets from HashiCorp Vault?

Yes, vault_read_secret and vault_list_secrets are low-risk read operations scoped by whatever policy the Vault token OrchestrAI uses carries. Reads are recorded in the replayable run like any other operation.

What happens if I ask OrchestrAI to revoke a Vault token or delete a policy?

Both are critical operations. OrchestrAI shows the target and requires a typed risk phrase before it proceeds, because revoking a token or removing a policy can cut off running services.

How does OrchestrAI authenticate to HashiCorp Vault?

You add a HashiCorp Vault credential once in the connections screen. It is envelope-encrypted with a per-record key wrapped by a cloud KMS and is only decrypted inside the run that needs it.

Integration · Secrets

HashiCorp Vault + OrchestrAI

Catalog exported 2026-09-02 · HashiCorp Vault website

Manage HashiCorp Vault secrets, policies, tokens, and auth methods from chat with confirmation on every write.

OrchestrAI exposes 14 HashiCorp Vault operations: 6 are low-risk (read-only or low-impact), 4 create or modify resources and run only after you confirm the plan, and 4 are destructive and require a typed risk phrase. 8 of them also carry a step-level approval gate.

14operations

6low risk

4create or modify

4destructive

8step-level approval

What teams use it for

Security and platform engineers use OrchestrAI to check what is stored under secret/data/payments without leaving the incident thread, write or rotate a secret, and review what a given policy actually grants before approving an access request. Token creation and revocation, policy deletion, and disabling an auth method are rated critical because each can lock people out or grant access, so they require a typed risk phrase. Enabling auth methods and creating or deleting secrets and policies are high risk and wait for confirmation. Secret engines, leases, seal status, and audit devices are not exposed, so mounting an engine or unsealing a node still happens through the Vault CLI.

Every HashiCorp Vault operation, with its risk level

HashiCorp Vault operations available through OrchestrAI
Operation	What it does	Risk	Step-level approval
`List Vault Auth Methods`	List enabled authentication methods	Low risk	No
`List Vault Policies`	List policies	Low risk	No
`List Vault Secrets`	List secrets at a path	Low risk	No
`Lookup Vault Token`	Lookup token info	Low risk	No
`Read Vault Policy`	Read a policy	Low risk	No
`Read Vault Secret`	Read a secret	Low risk	No
`Create Vault Policy`	Create or update a policy	Modifies existing	Yes
`Create Vault Secret`	Create or update a secret	Modifies existing	Yes
`Delete Vault Secret`	Delete a secret	Modifies existing	Yes
`Enable Vault Auth Method`	Enable an authentication method	Modifies existing	Yes
`Create Vault Token`	Create a new token	Destructive	Yes
`Delete Vault Policy`	Delete a policy	Destructive	Yes
`Disable Vault Auth Method`	Disable an authentication method	Destructive	Yes
`Revoke Vault Token`	Revoke a token	Destructive	Yes

Risk tiers come from the catalog: low is read-only or low-impact, medium creates resources and is reversible, high modifies existing resources, destructive may lose data. Every plan that creates or changes resources is shown with its cost estimate and waits for your confirmation. Operations marked with a step-level approval pause again on their own step. Destructive operations require a typed risk phrase.

What you connect

A HashiCorp Vault credential (stored as vault). Connected-service tokens are envelope-encrypted with a per-record key wrapped by a cloud KMS.

Prompts that work

Read the secret at secret/data/payments/stripe and tell me which keys it contains
Create a policy called ci-readonly that allows read on secret/data/ci/* and nothing else
Look up token accessor 7f3a and revoke it, it was pasted into a build log

Before anything runs

Every mutation shows its plan, cost estimate, and blast radius, then waits for your confirmation. Destructive operations require a typed risk phrase. Credentials are minted per run through OIDC federation and discarded afterward; nothing you create here is invisible later, because every resource lands in the desired-state ledger where drift is detected and can be converged. Details on the security page.

Frequently asked questions

Can OrchestrAI read secrets from HashiCorp Vault?: Yes, vault_read_secret and vault_list_secrets are low-risk read operations scoped by whatever policy the Vault token OrchestrAI uses carries. Reads are recorded in the replayable run like any other operation.
What happens if I ask OrchestrAI to revoke a Vault token or delete a policy?: Both are critical operations. OrchestrAI shows the target and requires a typed risk phrase before it proceeds, because revoking a token or removing a policy can cut off running services.
How does OrchestrAI authenticate to HashiCorp Vault?: You add a HashiCorp Vault credential once in the connections screen. It is envelope-encrypted with a per-record key wrapped by a cloud KMS and is only decrypted inside the run that needs it.

Related integrations

Argo CDGitOps HelmKubernetes KubernetesKubernetes

Try it on your own account

Connect your cloud read-only and see your resources, drift, and costs before anything runs. $5 minimum to start. Unused credits refunded in your first 14 days.

Start for $5

Unused credits refunded in your first 14 days.