SYSTEMS PERFORMANCE / Agency Agents community
WebAssembly boundary and runtime planner
Evaluates Wasm adoption through measured workloads, explicit memory ownership and capability-scoped runtime design.
“Expert WebAssembly engineer — compiling Rust/C++/Go to Wasm, JS interop and the boundary marshalling cost, WASI and server-side runtimes (Wasmtime/Wasmer), the component model, and near-native performance tuning.”
01 / THE REASONING
Why this made the selection.
- The wasm-bindgen example explicitly performs bulk copies and disclaims a zero-copy ABI; ownership is part of the boundary design.
- The workflow measures baseline, marshalling, instantiation and code size before adopting SIMD or threads.
- It treats repeated JS-to-Wasm calls as the usual reason a module looks slow, and it wants the loop inside the module with batched buffers.
- Server-side Wasm is framed as a deny-by-default WASI capability grant, with feature detection for SIMD, threads, and the component model.
02 / THE REVIEW RECORD
What we actually inspected.
Source review has boundaries.
A clear record is more useful than a “safe” badge.
Material inspected
- engineering/engineering-webassembly-engineer.md (complete definition)
- LICENSE (applicable redistribution terms)
- Host custom-agent documentation; repository installation guidance where applicable
Our findings
- The wasm-bindgen example explicitly performs bulk copies and disclaims a zero-copy ABI; ownership is part of the boundary design.
- The workflow measures baseline, marshalling, instantiation and code size before adopting SIMD or threads.
- The first rule says to move the loop into Wasm and cross the boundary with batched buffers.
- WASI guidance is to grant exact capabilities and no ambient host access. Feature detection is required where SIMD, threads, or the component model may be missing.
- Frontmatter does not set tools. The host inherit-all sentence applies if the file is loaded as a subagent.
- The profile does not register itself and does not build a module.
Not established by this review
- No agent was installed or invoked. Host/model compatibility, task accuracy and runtime behaviour were not tested.
- The resource was not executed and its tools were not called.
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.
- Read the pinned definition and MIT licence; review the host permissions before use.
- Read the full definition as a role reference; create a host-valid .claude/agents/ working copy and bound tools to the agreed task.
- Provide the non-Wasm baseline, representative buffers/workload, target browsers/runtime and allowed host capabilities.
Before you start
- A compatible agent host/model
- Project inputs and separately installed/authorized toolchains or services
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/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/engineering/engineering-webassembly-engineer.md
Supports: Workflow, inputs and outputs, Declared tools and constraints, Selection and limitations
https://github.com/msitarzewski/agency-agents/blob/bd956db032d907387e6b9ddebff3e18f1de8c679/LICENSE
Supports: Licence and attribution requirements
https://code.claude.com/docs/en/sub-agents.md
Supports: Host setup and permission defaults