Task Context Brief Template
Version: 1.0.0
Last updated: 2026-07-16
Purpose
Define the minimum complete, authorized context for one unit of model-assisted work.
Why
A task brief converts conversational intent into scope, evidence, acceptance criteria, and verification. It prevents assumptions from becoming hidden requirements.
How
<task_brief id="{{TASK_ID}}" version="{{VERSION}}">
<objective>{{OBJECTIVE}}</objective>
<scope>
<include>{{IN_SCOPE}}</include>
<exclude>{{OUT_OF_SCOPE}}</exclude>
</scope>
<inputs>
<source id="{{SOURCE_ID}}" revision="{{REVISION}}" trust="{{TRUST_CLASS}}">
{{AUTHORIZED_INPUT}}
</source>
</inputs>
<acceptance_criteria>
<criterion id="{{CRITERION_ID}}">{{MEASURABLE_CRITERION}}</criterion>
</acceptance_criteria>
<constraints>{{TASK_CONSTRAINTS}}</constraints>
<verification>{{VERIFICATION_COMMANDS_OR_REVIEW}}</verification>
<risk_and_approval>{{RISK_TIER_AND_APPROVER}}</risk_and_approval>
</task_brief>
Resolve ambiguities before high-impact execution. Do not request private chain-of-thought; request plans, evidence, and verification results as visible artifacts.
When
Use for implementation, review, research, data changes, and any handoff requiring reproducibility.
Tradeoffs
| Benefit | Cost |
|---|---|
| Clear completion contract | Authoring time |
| Bounded context | May require follow-up retrieval |
| Reproducible handoff | Brief can become stale |
Anti-Patterns
- “Do what we discussed” without source references.
- Unmeasurable acceptance criteria.
- Including unauthorized data.
- Approval requirements without a named authority.
Enterprise Considerations
Link briefs to change records, classify inputs, record approvers, and retain evidence according to the task's control domain.
Checklist
- All variables are resolved
- Scope and exclusions are explicit
- Inputs are authorized and revision-pinned
- Acceptance criteria are measurable
- Verification and approval are defined
Changelog
- 1.0.0 (2026-07-16): Initial task context brief template.
Version: AIES v1.0.0✏️ Edit this page on GitHub