CODE REVIEW / Anthropic
Rust API Ergonomics Reviewer
Reviews a Rust library from the downstream user’s side, and confirms a suspected lint with cargo before reporting it.
“Reviews Rust library/framework code from the downstream consumer's perspective. Focuses on API clarity, idiomaticity, happy-path friction, footguns, and what users will actually see in their editor and compiler output. …”
01 / THE REASONING
Why this made the selection.
- The role is limited to what a downstream Rust user types, sees in compiler output, and can misuse. Correctness bugs, performance and unsafe soundness are left to other review.
- A suspected clippy or check finding has to be confirmed with cargo check or cargo clippy. The default scope is git diff origin/main.
02 / THE REVIEW RECORD
What we actually inspected.
Source review has boundaries.
A clear record is more useful than a “safe” badge.
Material inspected
- .claude/agents/rust-api-ergonomics-reviewer.md
- LICENSE
- Claude Code custom-subagent location and capability documentation
- Cargo build-script lifecycle documentation
- Round 2: pinned definition, output contract and permission/dependency boundaries reread on 2026-10-14
Our findings
- The file says to report findings and not apply them, and to stop in two sentences when nothing above Low severity is present.
- High, medium and low are labels for findings, not a measured defect rate.
- The report is capped at 1500 words and groups user-visible problems by severity with file:line evidence; if nothing exceeds Low, it stops after two sentences.
- Checks distinguish downstream impl-site lints from allowances in the library workspace, and inspect generated-code ergonomics as well as handwritten public APIs.
- The checklist distinguishes struct-update defaults from non-exhaustive types that force mutate-after-default usage.
- It examines panicking builders for fallible siblings and doc-only invariants that could move into sealed traits or newtypes.
Not established by this review
- The agent was not executed; model behaviour, host compatibility and task outcomes were not runtime-tested.
The review applies to the material and revision named here. A newer upstream release can change its behavior.
03 / PUT IT TO WORK
Use the role in your project.
- Copy the reviewed file into the Claude Code agents directory your project already uses, or load it as a subagent from that path.
- Grant only Read, Glob, Grep and the git and cargo commands the file names. No installer was run.
Before you start
- Claude Code or a host that can load this Markdown role
- A Rust library checkout when the review is actually run
04 / FOLLOW THE EVIDENCE
The source trail.
Our notes are separate from the original resource.
Check upstream before adopting a new version.
https://github.com/anthropics/buffa/blob/5fed61b257d36dac48525a73c0fcb1e9fd8b5299/.claude/agents/rust-api-ergonomics-reviewer.md
Supports: summary, upstreamDescription, whySelected, limitations, access
- Apache-2.0 licence ↗Checked
https://github.com/anthropics/buffa/blob/5fed61b257d36dac48525a73c0fcb1e9fd8b5299/LICENSE
Supports: license
https://code.claude.com/docs/en/sub-agents
Supports: Project installation, Host tool and model boundaries
https://doc.rust-lang.org/cargo/reference/build-scripts.html
Supports: Cargo execution and output side effects