01
In-chat approval cards
Plans, document writes, knowledge writes, integration calls, MCP tools, and workflow runs pause until approve or reject.
Approval conceptsGovernance
What is Governance in Tale?
Governance is the seam between agent initiative and human judgement: approval cards hold actions until someone decides, every decision lands in the audit log, and spend stays visible — without a separate approver inbox.
How governance holds the line
Approve before actions ship, audit every run, compare models honestly, and keep agents inside your rules.
01 Automations
Every gated run leaves a log
A nightly access review runs on schedule and alerts #security on new admin grants — every execution numbered in the same journal approvals write to.
02 Arena
Compare before you commit
Policy-sensitive wording is exactly where models disagree. Arena drafts the same reply twice so the safer version ships.
03 Agents
Agents stay inside budgets and tools
Spend controls and tool grants travel with each agent — governance is not a separate bolt-on.
04 Knowledge
Cite what policy allowed the reply
Grounded answers show the policy and runbook sources so auditors can follow the same trail.
What does governance cover?
Put a human in the loop where it matters — with an audit trail, budget controls, and filters that keep sensitive data out of prompts and replies.
02
Suggest changes on integrations
Describe what is wrong; the agent revises the call instead of abandoning the work.
Approval concepts03
Audit trail
Each decision records actor, action, and timestamp. Resolved cards stay in the transcript.
Audit logs04
Guardrails and policies
Org-level guardrails and the run-code policy set the limits on what agents may attempt — before a card ever appears.
Guardrails05
PII filtered in both directions
Every chat message passes content safety, PII detection, and an optional moderation provider — on the way in and the way out. Matches block, mask, or flag; the matched text itself is never stored.
Guardrails06
Usage visibility
Usage analytics and budgets keep model spend visible to Admins.
Governance FAQ
Tale Governance holds agent actions behind in-chat approval cards, writes decisions to the audit log, and keeps usage visible to Admins.
No. Approvals are decided in the conversation they interrupt, by whoever holds that chat.
Yes. Approvals are configurable per action type — plans, document and knowledge writes, integration calls, MCP tools, and workflow runs can each require a human decision before they run.
Not if you require approval for it. Outbound actions hold as in-chat cards until someone decides — you choose which action types need a decision.
Keep processing on infrastructure you control, let PII detection mask personal data before it reaches or leaves a model, and point auditors at the audit log. Tale is GDPR-compliant, certified to ISO 27001 and SOC 2 Type II, and publishes its DPA and technical and organizational measures.
See Tale on your stack
Book a guided demo or talk to the team about self-hosted deployment, pricing, and hardware.