编程开发
代码生成、调试优化、技术文档等编程辅助提示词
675 个提示词
nullability, raises/throws, thread-safety, and a usage example for public API. Mark…
Docstrings/JSDoc: summary line (imperative), params/returns with units and
either (a) a rationale (constraint, tradeoff, external bug link, invariant) or (b) a…
Remove/replace comments that restate code; convert every surviving comment into
Turns bad comments into maintainable docs: explains WHY, keeps docs near code, and generates API docs from real signatures — no marketing fluff.
Code Comment / Documentation Improver
latency of replacing joins; report pipelines. Give a timeline with milestones.…
Risks: distributed transactions avoided via saga/eventual consistency;
parity (dual-run diffing), and cutover/rollback criteria.
Strangler steps as independently shippable PRs with feature flags, metrics for
dual-write/outbox → backfill → switch reads → stop old writes → drop columns, with…
Data ownership migration: read model/API replacements for foreign readers,
service — justify whether a service is even needed now (YAGNI) using team…
Target architecture: in-process module first (seams, interfaces), then optional
shared code, events), ranked by extraction difficulty.
Dependency analysis: inbound/outbound couplings (sync calls, shared DB joins,
invariants, and what other contexts currently read/write that data illegally.
Bounded-context definition: domain language, owned data (exact tables/columns),
Plans extraction of a bounded context from a monolith with data ownership and dual-write/cutover safety.
Monolith-to-Module / Service Extraction Planner
tests for changed lines (diff coverage), and why a global % target would be gamed.…
Recommend a realistic coverage policy: critical-package threshold + mandatory
clear names; show which production bug class each prevents.
Write the top 5 missing tests (boundary + failure paths), table-driven, with
if the feature were deleted, duplicated setup hiding intent, over-mocked units.
Identify WEAK tests: no assertions, assertion-free mocks, tests that would pass
trivial getters. Ignore generated code and DTOs.
Rank gaps by risk: money/auth/data-loss/permissions > complex branching >
integration paths are verified vs unverified. Produce a behavior coverage matrix.
Map behaviors (not lines) to tests: which business rules, branches, and
Identifies untested critical paths (not line coverage) and writes the highest-value missing tests first.
Test-Coverage Gap Analyzer
confidence; cite file:line for every claim. Ask me for specific files when needed.…
Confusing/dead-looking areas and questions to ask the team — label your
add job, add permission, change schema) with the test locations.
"Where would I change X?" map for 5 common changes (add endpoint, add field,
definitions appear.
Glossary of domain terms as used in the code, and where the authoritative
flags are typically done — with file references as examples.
Conventions: how errors, config, auth, logging, tests, migrations, and feature
the dependency rule the code seems to follow (or violate).
Draw an ASCII diagram of components and data flow; mark module boundaries and
prove no behavior change via tests. Provide the updated tsconfig/mypy config and a…
Submit in small commits: enable stricter rule per directory, fix violations,
hand-writing; add runtime validation and fail clearly on schema drift.
For JSON/API responses, generate types from the OpenAPI/JSON schema instead of
boundary responses, Record , broad exception types, optional fields used without checks.