Open source · Python · action-level runtime governance
For AI governance, model risk, and compliance
Turn policy into runtime evidence.
Show which policy applied to an agent action, why it was permitted or blocked, when a person was asked to decide, and what evidence remains for oversight.
5D is an action-policy gate for AI agents: it evaluates declared tool actions before the host executes them.
5D provides a versioned deterministic runtime-control layer. When an integration supplies the required action identity and outcome data, release-specific decision records can support reconstruction of declared governed actions.
Connect policy intent to a release-specific decision record when the integration supplies the required identity and outcome data, without confusing technical evidence with a legal or conformity conclusion.
The records show the properties tested, the assumptions made, and the limitations that remain. The integrating organisation retains responsibility for identity, execution, outcome capture, path coverage, and any legal or assurance conclusion.
Relevant deployment example
Follow one action relevant to your role.
Synthetic tested action under the pinned 5D 0.7.0 illustrative policy.
Travel service agent
Send a policy exception and refund promise
Generated guidance contradicts the applicable policy and appears authoritative. Under the pinned illustrative policy, the expected result is to hold the action until the host records the required review.
D2T2R3E3A3
Hold for review · Score 1.94
Tool
message.send
Target
individual traveler
Purpose
resolve request
5D suppliesAn ORANGE hold signal before sending the commitment.
Host ownsPolicy retrieval, content review, reviewer authority, and final send.
Coverage limit5D does not determine the correct refund policy.
Your evaluation lens
Follow how governance intent becomes a decision, where human oversight applies, and which evidence remains available later.
01
Traceability
Versioned policy context
Show which active policy and configuration governed the proposed action.
02
Human-review evidence
Meaningful human oversight
Route only policy-selected cases to a reviewer and preserve who decided what, when, and why.
03
Oversight evidence
Evidence mapping
Map runtime records to oversight concepts without presenting a legal or conformity assessment.
04
Change control
Governed change
Replay outcomes and promote policy changes through explicit, versioned review.
Evidence your role can use
Know which evidence 5D produces, and which evidence it does not.
Connect policy provenance and action records to oversight without turning technical evidence into a compliance conclusion.
Runtime record
Decision record
It lets teams reconstruct which policy path evaluated one proposed action and what 5D decided.
Policy provenance
Policy digest
It distinguishes the policy content actually used, even when a human-readable version label is reused.
Integrity result
Chain check
It can reveal a break in the local chain over the decision-time fields that 5D covers.