agentleFS
NoorQureshiNoorQureshi/SploitAgent/skills/code-review/code-review-rust/SKILL.md

code-review-rust

Skill26 starsChanged 31 days agoMITView on GitHub

Security review of Rust code — where a memory-safe language still has real bugs: `unsafe`, FFI, panics, and the usual injection/logic sinks. Load when reviewing Rust source/PR, on .rs / Cargo.toml in scope, or "review this Rust". Signals: Cargo.toml, unsafe blocks, extern "C", unwrap/expect, actix/axum/rocket handlers.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    unsafe blocks: every one is a manual proof obligation — raw pointer deref, get_unchecked, slice::from_raw_parts, mem::transmute, …

    line 31
  2. Very specific:

    Run cargo audit (known-vuln deps), cargo clippy, and semgrep.

    line 47

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Rust source review

When it applies

Reviewing Rust (a service, a CLI, a library). Safe Rust removes most memory bugs, so the review focuses on the places safety is opted out of, the ways Rust code still panics or mis-handles input, and ordinary injection/logic in web handlers.

Why it works

Rust's guarantees hold only outside unsafe and only for memory safety — they don't stop SQL built by string, a command run via a shell, a .unwrap() that panics on attacker input (DoS), or a logic error. Concentrating on those boundaries finds the real bugs efficiently.

Sinks & patterns (grep, then reason about the boundary)

  • unsafe blocks: every one is a manual proof obligation — raw pointer deref, get_unchecked, slice::from_raw_parts, mem::transmute, uninitialised memory. Check the invariant it assumes.
  • FFI: extern "C"/bindgen boundaries — lengths, lifetimes, and NUL handling across the C edge (the C side has none of Rust's guarantees; pair with code-review-cpp).
  • Panics as DoS: unwrap/expect/panic!/indexing v[i]/unreachable! on attacker-controlled input; integer as casts that truncate; arithmetic overflow (panics in debug, wraps in release — both can be bugs). Prefer ?/checked ops.
  • Injection: SQL via format! into a query instead of parameter binding (sqlx/diesel); std::process::Command with sh -c and concatenated input; Command arg vs shell form.
  • Path / SSRF: user paths joined without canonicalisation + prefix check; HTTP clients (reqwest) fetching user URLs.
  • Deserialization / web: serde into types from untrusted data (resource exhaustion, unexpected variants); actix/axum/rocket extractors bound to over-broad structs (mass assignment); missing auth middleware on state-changing routes.

Method

  1. Run cargo audit (known-vuln deps), cargo clippy, and semgrep; miri for unsafe UB where feasible.
  2. rg 'unsafe|unwrap\(\)|expect\(|transmute|Command::new|format!\(.*(SELECT|INSERT|UPDATE)' → review each.
  3. Justify every unsafe block's invariant; if you can't, that's a finding.
  4. Confirm exploitable injection/logic with the matching runtime skill.

Gotchas

  • Idiomatic Rust is genuinely safe — don't invent memory bugs in safe code; spend effort on unsafe, FFI, panics, deps, and logic.
  • Release-mode integer overflow wraps silently — a checked_*/saturating_* audit matters for size/index math.
  • cargo audit flags vulnerable crates you'd never see by reading — always run it.

References

Rustonomicon (unsafe); RustSec advisory DB / cargo-audit; Clippy lint set; Secure Rust Guidelines (ANSSI).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.