API Key Management
Establish programmatic access capabilities in the order of confirming call scope, creating keys, tightening permissions, and API validation.
Feature Overview
API Keys provide an identity for programs, scripts, and external systems to access Clouisle. They are not account substitutes for human login, but an independent way to govern program identities.
Applicable Scenarios
API Keys are commonly used for:
- External systems calling Agents or workflows
- Automation platforms accessing knowledge bases, files, or conversation APIs
- Scheduled tasks, bots, or integration platforms calling APIs long-term
Prerequisites
Before you start, we recommend confirming:
- Who the calling system is
- Which API scopes need to be accessed
- Whether expiration time, enable or disable control, and rate limiting are needed
- Whether a minimum calling sample has been prepared
Operation Steps
Step 1: First Clarify the Key's Usage Scope
Before creating it, clearly answer what this Key is used for, such as:
- Only calling a specific Agent
- Only triggering a specific workflow
- Needing access to multiple business capability APIs
The clearer this step is, the easier it is to tighten permissions later.
Step 2: Create the Key and Use a Traceable Naming Method
When creating it, we recommend the name include at least:
- Calling system name
- Environment, such as test or production
- Purpose description
For example: crm-prod-agent-call.
This makes it easier to quickly identify later in logs, audit, or rotation.
Step 3: Limit Permissions and Lifecycle
After creation, focus on confirming:
- Which resources are allowed to be accessed
- Whether it is enabled by default
- Whether an expiration time is set
- Whether rate limiting is configured
In production environments, do not open full permissions at the beginning just for convenience.
Step 4: Immediately Perform One Minimum API Validation
After the Key is created, immediately run one minimum test with a real request and confirm at least:
- Authentication succeeds
- The target API can be called
- Unauthorized APIs are rejected
If this step is not validated, it is easy to mistake permission problems for API problems during later joint debugging.
Step 5: Add Disable, Rotation, and Reclamation Rules
After the Key starts entering official systems, also define:
- How long before expiration it should be rotated
- Who reclaims it after the system is decommissioned
- How to urgently disable it if abnormal leakage occurs
Result Validation
A properly governed API Key should at least satisfy:
- Target API calls succeed
- Unauthorized API calls fail
- It becomes invalid immediately after being disabled
- Last usage time and purpose ownership can be tracked
FAQ
Why Do API Key Calls Keep Failing
Check first:
- Whether the Key is valid
- Whether the request header is correct
- Whether the called API exceeds the authorized scope
- Whether the environment address is wrong
Why Is It Not Recommended for Multiple People or Systems to Share the Same Key
Because once a problem occurs, it is difficult to trace responsibility and difficult to separately revoke one system's access permission.
Why Set Expiration Time and a Rotation Mechanism
If a Key remains valid long-term and no one manages it, once it leaks, the risk will continue to exist. Rotation and reclamation mechanisms are indispensable parts of program identity governance.
Notes
- One Key per system, issued separately by environment
- API Keys should always follow the principle of least privilege
- Before official joint debugging, first use a minimum request to verify whether authentication and authorization are correct
Roles and Permissions
Establish a maintainable authorization system in the order of role modeling, permission mapping, and regression validation with real accounts.
API Key Usage Practices
Establish stable API Key usage standards from naming, authorization, validation, rotation, to long-term governance.