Tools · Planning and checking
decision_matrix
Read-onlyCompare options against explicit criteria and recommend the strongest choice.
When to use: Use this when there are multiple valid paths and the agent should recommend one clearly.
Example
Run for real by scripts/tool-examples.py in a small demo project, with Haiku 5.5 where the tool calls a model. The output is exactly what the tool returned.
The call
{
"tool": "decision_matrix",
"arguments": {
"decision": "Which job queue for exports?",
"options": [
"Celery",
"RQ",
"Dramatiq"
],
"criteria": [
"operational cost",
"retries and backoff",
"community"
]
}
}Real output
Decision: Which job queue for exports? Comparison Matrix: - Celery: evaluate against operational cost, retries and backoff, community - RQ: evaluate against operational cost, retries and backoff, community - Dramatiq: evaluate against operational cost, retries and backoff, community Recommendation: Celery Fallback: RQ Tradeoffs: - Favor the option with the strongest fit to the highest-priority criteria. - Keep a fallback when execution risk is high.
Parameters
| Name | Type | Description |
|---|---|---|
| decision* | string | Decision to make |
| options* | array | Candidate options |
| criteria* | array | Evaluation criteria |
* required
More in Planning and checking
Plan the work, weigh options, and check results before answering.