ValentinNikolaev/agent-skills / agent-plugins/skills/review-spec
review-spec
VerifiedReview proposed specifications for feasibility, gaps, and testability; not plans or code conformance.
| File | Kind | Size |
|---|---|---|
| LICENSE.sergeyklay-agents | other | 11.1 KB |
| LICENSE.spf13-go-skills | other | 1.0 KB |
| NOTICE.sergeyklay-agents | other | 965 B |
| SKILL.md | skill md | 8.9 KB |
| agents/openai.yaml | config | 235 B |
| references/go-spec-checks.md | markdown | 3.3 KB |
| references/review-cost-examples.md | markdown | 2.8 KB |
references/go-spec-checks.md
View on GitHub# Go Specification Checks
Use these checks for a proposed Go design, not as a second code review or an implementation plan. Establish the project's `go.mod`, toolchain policy, existing contracts, and architecture decisions before applying an idiom; reuse supplied research/shared context under the parent skill's evidence rules instead of re-reading established facts. Check whether package and ownership obligations are explicit, leaving competing architecture designs to `review-arch`.
## Contracts and failure paths
- Identify consumers of proposed interfaces and the capability each consumer needs. Prefer a concrete type or small consumer-owned interface when it meets the requirement, but do not reject an established abstraction solely because it has one current implementation.
- Specify error outcomes, caller-visible error identity, partial results, and retry/idempotency behavior where they affect the contract. Do not require exposing an internal database or transport error merely to use `%w`.
- State ownership and cleanup for connections, streams, files, transactions, and background work. Specify cancellation, shutdown, outstanding work, and the bound on fan-out or queued input when relevant.
- Check that proposed APIs exist in the supported Go version. `WaitGroup.Go` does not itself bound concurrency or propagate errors. Separate a toolchain upgrade decision from the proposed feature.
- Evaluate package dependencies and reuse against actual responsibilities. `internal/`, nested packages, worker pools, generics, and architectural styles are not defects by name; identify a concrete dependency, lifecycle, or complexity cost.
## Cobra and Viper, only when used
- Identify where new commands are registered and how command state is isolated between invocations and tests. Factory construction is useful when mutable global commands would leak state; a mandatory rewrite of unrelated commands is not a spec requirement.
- Distinguish syntax/argument failures from runtime failures. Specify error propagation, exit status, output streams, and cancellation rather than requiring a particular handler method without examining its behavior.
- Specify flag/config/environment precedence and required versus optional configuration. If Viper unmarshals a struct, ensure environment-only keys are registered with defaults or explicit bindings; `AutomaticEnv` alone does not guarantee discovery of every struct field.
- For file-writing commands, address equal input/output targets, overwrite rules, partial output, and cleanup where data loss is possible.
- Require observable acceptance cases for parsing, configuration, errors, and cancellation. Exercise the real root-to-subcommand path when registration or inherited hooks matter.
## Report
Tie every concern to a spec section and its consequence. Keep harmless preferences out of blockers; put unresolved material assumptions in Questions. Use the parent skill's `READY`, `REVISE`, or `BLOCKED` verdict. Do not edit the spec, invoke subagents automatically, or add implementation details the design deliberately leaves open.
Adapted and substantially rewritten from [spf13/go-spec-reviewer](https://github.com/spf13/go-skills/blob/9ac6eca43161163bb21621a520a57df50d5ad464/go-spec-reviewer/SKILL.md), under the bundled `LICENSE.spf13-go-skills`. Host-specific paths, automatic delegation, and universal architecture prescriptions are omitted.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirements | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |