AGENT SKILL

rust-ffi-safe-wrappers

Design and review Rust FFI boundaries that keep raw declarations isolated behind safe ownership, lifetime, error, and cleanup wrappers. Use when writing, refactoring, or reviewing unsafe extern blocks, C ABI wrappers, repr(C) types, CString, CStr, raw handles, Drop cleanup, bindgen, cbindgen, or panic-safe foreign boundaries.

License 许可
MIT
最近更新
2026-11-14
Stars
★ 1.6k
Forks
491

01 前言 Front Matter

name
rust-ffi-safe-wrappers
license
MIT
metadata
{ "tags": "rust,ffi,c-abi,unsafe,bindings", "version": "0.1.0", "category": "Rust", "displayName": "Rust FFI Safe Wrappers" }
description
Design and review Rust FFI boundaries that keep raw declarations isolated behind safe ownership, lifetime, error, and cleanup wrappers. Use when writing, refactoring, or reviewing unsafe extern blocks, C ABI wrappers, repr(C) types, CString, CStr, raw handles, Drop cleanup, bindgen, cbindgen, or panic-safe foreign boundaries.
compatibility
Codex, Claude Code, and other Agent Skills-compatible clients.

02 资源文件 2

03 SKILL.md 正文

Rust FFI Safe Wrappers

Use this skill to wrap foreign interfaces in Rust APIs that make ownership, lifetimes, errors, and cleanup explicit. FFI code should be unsafe at the raw boundary and safe for normal Rust callers.

Core Workflow

  1. Separate raw declarations from safe wrappers. Put raw unsafe extern bindings in a small private module.
  2. Verify ABI, type layout, calling convention, ownership, nullability, threading, and error conventions against the foreign header or docs.
  3. Use unsafe extern "C" blocks for modern Rust, especially Rust 2024 edition code. Leave safety-conditional items unqualified or explicitly unsafe, and mark individual extern items safe only when calling them is safe for all Rust inputs.
  4. Use #[repr(C)] for structs and enums that cross the ABI boundary. Do not expose Rust-only layout types across C.
  5. Use CString for owned nul-terminated strings sent to C and CStr for borrowed C strings. Never treat arbitrary C memory as Rust-owned.
  6. Wrap raw handles in Rust types with private fields and Drop cleanup using the matching foreign deallocator.
  7. Prevent Rust panics from crossing extern "C" boundaries. Catch panics or expose an ABI that permits unwinding only when deliberately using "C-unwind".

Wrapper Rules

Read references/ffi-boundary-patterns.md before adding or reviewing FFI bindings.

  • Prefer core::ffi or std::ffi C types over guessing integer sizes.
  • Convert foreign error codes into Rust Result at the wrapper boundary.
  • Represent nullable handles as Option<NonNull<T>> internally when useful, but expose safe Rust types to callers.
  • Tie borrowed values to owner lifetimes with normal references or PhantomData when the compiler cannot see the relationship.
  • Make initialization and shutdown idempotent with OnceLock, LazyLock, or Once when the foreign library requires process-wide setup.
  • Use bindgen for large or changing C headers and cbindgen when exporting a Rust API to C, but still review generated unsafe signatures.

Rust 2024 Syntax Checks

unsafe extern "C" {
    pub fn library_open(path: *const core::ffi::c_char) -> *mut RawHandle;
    pub fn library_close(handle: *mut RawHandle);
    // Mark an item `safe` only when calling it is valid for all Rust inputs:
    // pub safe fn library_version() -> core::ffi::c_int;
}

// SAFETY: This exported symbol name is unique for this library.
#[unsafe(no_mangle)]
pub extern "C" fn plugin_version() -> core::ffi::c_int {
    1
}

Unsafe attributes such as no_mangle, export_name, and link_section need the unsafe(...) form in Rust 2024 edition code. Include a SAFETY comment where symbol uniqueness or linker behavior is part of the contract.

Review Checklist

  • Raw bindings are private or clearly separated from safe wrappers.
  • Every pointer parameter has a documented nullability and ownership rule.
  • Every allocation is freed by the same side or an explicitly matching deallocator.
  • Strings and buffers preserve length, encoding, and nul-byte requirements.
  • Panics, callbacks, and threads crossing the boundary have explicit behavior.

04 同仓库更多 Skills

hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-review-with-annotations

Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-find-and-synthesize-context

Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-update-docs

Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-manage-records

Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-create-or-manage-interactive-html

Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

worktable-collaborate-threads

Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.

★ 1.6k