메뉴
Use when writing MASM `u32*` instructions on values from user input or untrusted sources — ensure the operands are valid u32s first.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use when writing kernel, account, or note MASM code that reads from or writes to the advice provider (advice stack / advice map) — validate advice data.
Use when writing or reviewing MASM hot paths — prefer the cheaper equivalent instruction: `neq.0` over `push.0 gt` for non-zero checks, `cdrop` over an `if/else` selecting between two values, `dup.N` over `loc_load` for a value still on the stack, `eqw` over hand-rolled element-wise word comparison, `u32gt`/`u32lt` over generic `gt`/`lt` on known-u32 operands.
Use when writing a GENERIC MASM storage utility (one that operates over a caller-chosen slot or map) inside a reusable account component — receive the slot id as a parameter so the utility works against any slot, not one hard-coded one.
Use when constructing a `Felt` from a numeric value in Rust — avoid silently producing a non-canonical Felt that no longer equals the original input.
Critical pitfalls and safety rules for Miden frontend development. Covers WASM initialization, concurrent access crashes, COOP/COEP headers, BigInt handling, Bech32 network mismatches, IndexedDB state loss, auto-sync side effects, Vite configuration, and React rendering race conditions. Use when reviewing, debugging, or writing Miden frontend code.
Guide for advanced Miden frontend development using source repo exploration. Covers AI development practices (Plan Mode, verification-driven development, context engineering, sub-agents) and maps the Miden web-sdk source repository for discovering advanced patterns. Use when building complex applications beyond basic hook usage, implementing custom signers, working with WasmWebClient directly, or troubleshooting SDK internals.
| name | u32-assert-before-u32-ops |
| description | Use when writing MASM `u32*` instructions on values from user input or untrusted sources — ensure the operands are valid u32s first. |
MASM's u32* instructions require their operands to be valid u32 values (i.e. fit in 32 bits, <= u32::MAX). Applied to a non-u32 value the behavior is undefined: in the current VM such an op typically traps with the generic error (operation expected u32 values, but got values: ...) rather than telling you which precondition was violated — but trapping is not guaranteed (it may instead wrap/truncate and silently poison the proof).
Before applying any u32* instruction to a value that is not already known to be a valid u32 (e.g. it came from the stack as input, was read from memory, or arose from a non-u32 arithmetic op), assert the bound with a descriptive error:
u32assert # assert the one value on top of the stack
u32assert2 # assert the two values on top of the stack
u32assertw # assert the four elements of the word on top of the stack
All three accept a named error via .err=, e.g. u32assert2.err=ERR_NOT_U32.
If the operand is already known-valid (just produced by another u32* op, or a value loaded from a slot whose layout is u32 by construction), skip the assert.
u32* instructions are tuned for the precondition that operands fit in 32 bits. Passing an out-of-range operand to a u32* op is undefined behavior: the current VM almost always traps with the generic operation expected u32 values, but got values: ... error, but trapping is not guaranteed — it may instead wrap/truncate and silently poison the proof, and even when it traps it does not name the precondition or the call site. Asserting first with u32assert*.err= turns this fragile/undefined path into a deterministic, named failure mode, so a non-u32 input fails loudly and diagnosably at the point where the assumption is introduced.
# Good: assert u32 before the u32 op
u32assert.err=ERR_VALUE_NOT_U32
u32overflowing_add
# Good: both operands at once
u32assert2.err=ERR_VALUES_NOT_U32
u32lt
# Bad: u32 op on untrusted input
u32overflowing_add # if an operand exceeds 2^32 the behavior is undefined; the current
# VM typically traps with the generic "operation expected u32 values"
# error instead of a named one (but may instead wrap and poison the proof)