Explore Prompts

Create a rule-based + learned risk classifier for diffs: file types, touched modules, dependency impact, and security sensitivity. Use it to decide required gates.

Design an agent that updates docs alongside code changes: API docs, READMEs, and changelogs. Include a doc quality checklist and link validation.

Create an agent to optimize CI time: caching, test sharding, and selective runs while preserving safety. Include a policy to prevent skipping critical tests.

Design an agent that identifies flaky tests, isolates root causes (timing, ordering, external deps), and proposes stabilization patches. Require proof via repeated runs.

Define error budgets and a policy: when error budget is low, block auto-deploys and require human approval. Include dashboards and alert thresholds.

Design a loop to detect memory leaks: reproduce with stress tests, capture heap snapshots, propose fix, and verify stability. Include regression tests.

Create a performance tuning agent that profiles hotspots, proposes targeted optimizations, validates with benchmarks, and documents tradeoffs.

Design a migration workflow: generate migrations, run on staging snapshots, verify invariants, measure runtime, and define rollback strategy. Require human approval for prod.

Create an agent that checks API changes for backward compatibility: OpenAPI/JSON schema diffs, deprecation policy, versioning, and migration notes.

Design characterization tests for legacy code before refactoring: capture I/O behavior, edge cases, and performance baselines. Use them as a gate for changes.

Create an agent that infers code style from the repo (naming, patterns, architecture) and generates a style guide. Use it to validate future edits.

Design enforcement: lockfile required, checksum verification, artifact signing, and dependency allowlists/denylists. Include alerting on suspicious updates.