Curioprompt
Add Explore New Best

Prompt Cards

Model Selection Strategy: Strong Model vs Cheap Model
Create a model selection strategy: route tasks by complexity, use smaller models for substeps, and add verifier checks. Include quality/cost tradeoffs and metrics.
Tags: model-routing, cost, quality, verification, optimization
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Self-Improvement Loop: Safe Prompt/Tool Iteration
Design a safe self-improvement loop for prompts/tools: propose change, run evals, red-team, canary, deploy, monitor. Include stop rules and audit artifacts.
Tags: self-improvement, iteration, safety, evals, canary
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Plan Quality Rubric: What ‘Good Plans’ Look Like
Create a plan quality rubric: completeness, dependency ordering, risk controls, verification steps, and rollback. Provide examples of high-quality vs low-quality plans.
Tags: planning, rubric, quality, verification, checklists
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Agent UX: Ask Clarifying Questions Intelligently
Design a questioning strategy: when to ask, how to propose defaults, and how to avoid excessive back-and-forth. Include a “progress-first” interaction model.
Tags: UX, clarification, defaults, interaction-design, agents
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Policy Profiles: Different Safety Levels by User Type
Create policy profiles for user tiers: consumer, enterprise, regulated. Include stricter tool permissions, logging requirements, and human approval gates by tier.
Tags: policy-profiles, enterprise, regulated, permissions, governance
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Multi-Tenancy Design: Enterprise-Grade Isolation
Design an enterprise multi-tenant architecture: per-tenant tool scopes, data isolation, policy profiles, and audit reporting. Include threat model and testing plan.
Tags: enterprise, multi-tenancy, isolation, policy, audit
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Secure Secrets Handling for Tool-Using Agents
Design secrets handling: vault integration, short-lived tokens, environment isolation, and audit trails. Include tests to ensure secrets never appear in outputs or logs.
Tags: secrets, security, vault, audit, containment
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Privacy-by-Design: Minimal Data, Maximum Utility
Design a privacy-by-design approach: data minimization, redaction, retention limits, and access controls for agent logs and memory. Include a compliance-friendly checklist.
Tags: privacy, data-minimization, redaction, retention, security
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Agent Runbooks: On-Call Playbook for Failures
Create operational runbooks: common failures, triage steps, rollback, and user comms. Include SLO breaches, tool outages, and prompt regressions.
Tags: runbooks, ops, on-call, incident-response, reliability
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Safety Case Template for a New Agent Capability
Write a safety case template: intended use, hazards, mitigations, tests, monitoring, and rollback. Include what evidence is required to ship the capability.
Tags: safety-case, governance, testing, monitoring, release
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Reasoning Transparency: Explain Decisions With Evidence
Create a policy for explanations: cite sources, show assumptions, and label uncertainty. Provide UI patterns for “why this tool was called” and “why this conclusion.”
Tags: transparency, explanations, uncertainty, citations, trust
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View
Artifact-Centric Agents: Make Work Inspectable
Design an artifact-centric workflow: agents produce plans, checklists, diffs, and reports as first-class outputs. Include templates and how artifacts enable auditing and reuse.
Tags: artifacts, auditability, workflow, templates, reuse
Author: Assistant
Created at: 2026-01-06 00:00:00
Average Rating:
Total Ratings:
View

Curio AI Brain

Available in Chrome Web Store!