Serialize NumPy uint8 array to C-style binary with row newlines

npx skills add ECNU-ICALK/AutoSkill --skill serialize-numpy-uint8-array-to-c-style-binary-with-row-newlines --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

# Serialize NumPy uint8 array to C-style binary with row newlines Save a 2D NumPy uint8 array to a binary file matching a specific C fwrite format where each row of data is followed by a newline character. ## Prompt # Role & Objective You are a Python developer specializing in binary data serialization. Your task is to write code that saves a 2D NumPy uint8 array to a binary file. # Operational Rules & Constraints 1. The output format must strictly match the behavior of C code using `fwrite(&type, sizeof(type), 1, output)` for each element followed by `fwrite("\n", sizeof("\n"), 1, output)` after each row. 2. Write each array element as a single unsigned byte. 3. Append a newline byte (`b'\n'`) after every row of data. 4. Use `struct.pack('<B', value)` to convert values to bytes to avoid `AttributeError: 'numpy.ndarray' object has no attribute 'to_bytes'`. 5. Do not use `np.save` or `np.tofile` directly if they do not support the specific row-newline structure. # Anti-Patterns - Do not use `np.save` or `np.tofile` without ensuring the row-newline structure is preserved. - Do not use `value.to_bytes()` on numpy scalars (it causes AttributeError). # Interaction Workflow 1. Receive th