code-review

Files (2)

• SKILL.md
• openai.yaml · agents/

SKILL.md (first lines)

Two-axis review of the diff between `HEAD` and a fixed point the user supplies:

- **Standards** — does the code conform to this repo's documented coding standards?
- **Spec** — does the code faithfully implement the originating issue / spec?

Both axes run as **parallel sub-agents** so they don't pollute each other's context, then this skill aggregates their findings.

### Runtime boundary

This workflow is runtime-neutral. Keep the output language driven by `language` from
`.harness/project.json`, and use the existing manually configured review mechanism. The workflow
must not launch a runtime-specific adapter, choose a provider or model, or change the reviewed
branch.

The issue tracker should have been provided to you. If `docs/agents/issue-tracker.md` is missing, tell the user to run `harness update`, which seeds it from the harness templates. On GitLab, every `glab` command takes `-R <project-url>`; `<project-url>` and the other GitLab placeholders (`<host>`, `<project-id>`) are defined in `docs/agents/issue-tracker.md` → GitLab → Conventions.

## Process

### 1. Pin the fixed point

Whatever the user said is the fixed point — a commit SHA, branch name, tag, `main`, `HEAD~5`, etc. If they didn't specify one, ask for it.

If the user names a PR or MR, its target branch is the fixed point. Read the target and source branches — GitHub: `gh pr view <n> --json baseRefName,headRefName`; GitLab: `glab mr view <iid> -R <project-url> -F json` (`target_branch`, `source_branch`). Then run `git fetch origin <target>` and use `origin/<target>` as the fixed point. The current checkout must be the source branch; if it is not, tell the user instead of switching branches (see _Runtime boundary_).

Capture the diff command once: `git diff <fixed-point>...HEAD` (three-dot, so the comparison is against the merge-base). Also note the list of commits via `git log <fixed-point>..HEAD --oneline`.

Before going further, confirm the fixed point resolves (`git rev-parse <fixed-point>`) and the diff is non-empty. A bad ref or empty diff should fail here — not inside two parallel sub-agents.

For the gated `/implement` route, review also consumes the coordinator's pinned Context Package and
candidate SHA. The coordinator must verify the integration base freshness before review and publish;
stale base is handled by a new developer/rebase dispatch. A delta-review is either test-only (after
a Warning: a new candidate with a test-only diff, brief field `delta_review_of`) or after a
fix-forward (brief section `delta_review_scope`, chosen by the coordinator). A fix-forward
delta-review reviews both axes on the delta alone and relies on the prior review's report for the
rest of the candidate. The coordinator escalates it to a full review when the candidate or the new
commits match a risk trigger the prior review did not see, change a file outside the carried items,
break a rebased copy's `git patch-id`, drop a previous-candidate commit, or add no commits. Either
delta-review is an independent review dispatch, never an in-place replacement of the original
evidence.

### 2. Identify the spec source

Look for the originating spec, in this order:

1. Issue references in the commit messages or the named PR/MR (`#123`, `Closes #45`, `Related to #45`, etc.) — fetch via the workflow in `docs/agents/issue-tracker.md`: GitHub `gh issue view <n> --comments`; GitLab `glab issue view <n> -R <project-url> --comments`. A GitLab `!67` is a merge request, not an issue: read it with `glab mr view 67 -R <project-url> --comments` (GitHub PR: `gh pr view <n> --comments`) and take its ticket from the `Closes #<ID>`/`Related to #<ID>` footer.
2. A path the user passed as an argument.
3. A spec file under `docs/`, `specs/`, or `.scratch/` matching the branch name or feature.
4. If nothing is found, ask the user where the spec is. If they say there isn't one, the **Spec** sub-agent will skip and report "no spec available".

### 3. Identify the standards sources

Anything in the repo that documents how code should be written, such as `CODING_STANDARDS.md` or `CONTRIBUTING.md`.

On top of whatever the repo documents, the Standards axis always carries the **smell baseline** below — a fixed set of Fowler code smells (_Refactoring_, ch.3) that applies even when a repo documents nothing. Two rules bind it:

- **The repo overrides.** A documented repo standard always wins; where it endorses something the baseline would flag, suppress the smell.
- **Always a judgement call.** Each smell is a labelled heuristic ("possible Feature Envy"), never a hard violation — and, like any standard here, skip anything tooling already enforces.

Each smell reads _what it is_ → _how to fix_; match it against the diff:

- **Mysterious Name** — a function, variable, or type whose name doesn't reveal what it does or holds. → rename it; if no honest name comes, the design's murky.
- **Duplicated Code** — the same logic shape appears in more than one hunk or file in the change. → extract the shared shape, call it from both.
- **Feature Envy** — a method that reaches into another object's data more than its own. → move the method onto the data it envies.
- **Data Clumps** — the same few fields or params keep travelling together (a type wanting to be born). → bundle them into one type, pass that.
- **Primitive Obsession** — a primitive or string standing in for a domain concept that deserves its own type. → give the concept its own small type.
- **Repeated Switches** — the same `switch`/`if`-cascade on the same type recurs across the change. → replace with polymorphism, or one map both sites share.
- **Shotgun Surgery** — one logical change forces scattered edits across many files in the diff. → gather what changes together into one module.
- **Divergent Change** — one file or module is edited for several unrelated reasons. → split so each module changes for one reason.
- **Speculative Generality** — abstraction, parameters, or hooks added for needs the spec doesn't have. → delete it; inline back until a real need shows.
- **Message Chains** — long `a.b().c().d()` navigation the caller shouldn't depend on. → hide the walk behind one method on the first object.
- **Middle Man** — a class or function that mostly just delegates onward. → cut it, call the real target direct.
- **Refused Bequest** — a subclass or implementer that ignores or overrides most of what it inherits. → drop the inheritance, use composition.

### 4. Review both axes

Launch the `code-review-standards` and `code-review-spec` subagents defined in `.claude/agents/` through the coding application's manually configured mechanism. Wait for both reports, then return them to this primary session for aggregation. Do not replace either axis with an inline review.

**`code-review-standards` prompt** — include:

- The full diff command and commit list.
- The list of standards-source files you found in step 3, **plus the smell baseline from step 3** pasted in full — the sub-agent has no other access to it.

**`code-review-spec` prompt** — include:

- The diff command and commit list.
- The path or fetched contents of the spec.

If the spec is missing, skip the `code-review-spec` sub-agent and note this in the final report.

### 5. Aggregate

Present the two reports under `## Standards` and `## Spec` headings, verbatim or lightly cleaned. Do **not** merge or rerank findings — the two axes are deliberately separate (see _Why two axes_).

End with a one-line summary: total findings per axis, and the worst issue _within each axis_ (if any). Don't pick a single winner across axes — that's the reranking the separation exists to prevent.

## Why two axes

A change can pass one axis and fail the other:

- Code that follows every standard but implements the wrong thing → **Standards pass, Spec fail.**
- Code that does exactly what the issue asked but breaks the project's conventions → **Spec pass, Standards fail.**

Reporting them separately stops one axis from masking the other.

### Output language

Read `language` from `.harness/project.json` (default `ru` if the file or field is absent). If `ru`, write the entire report — both sub-agent briefs and the aggregated output — in Russian regardless of the prompt's language. If `en`, write in English.

From PVMalove/claude-agent-harness, MIT.

More from PVMalove/claude-agent-harness