Academic Dataset Sanitization and Scrubbing Script

Role: Senior Research Software Engineer specializing in reproducible data pipelines.

Context

• Raw survey and experimental data format: {{source_data_format}}
• Explicit direct and indirect identifiers requiring removal: {{pii_fields_to_redact}}
• Target execution runtime and language: {{target_environment}}
• Institutional governance and compliance threshold: {{retention_policy_rules}}
• Volume and record batch size: {{dataset_size_scale}}
• Quality assurance and statistical integrity standards: {{validation_criteria}}

Task

Construct a standalone, idempotent data sanitization and anonymization script that ingests raw participant records, strips restricted attributes according to research governance standards, and outputs clean analytical datasets ready for peer-reviewed distribution.

Method

1. Parse input flags and configure stream buffers for {{dataset_size_scale}} processing.
2. Inspect record schemas to verify structural compatibility with {{source_data_format}}.
3. Apply deterministic hashing or zeroing functions to all targets identified in {{pii_fields_to_redact}}.
4. Implement pseudo-identifier generalization aligned with {{retention_policy_rules}} without distorting variance.
5. Execute pre-write validation checks against {{validation_criteria}} to guarantee schema completeness.
6. Generate an immutable cryptographic manifest containing input hashes and output record counts.
7. Write sanitized records safely to target destination using atomic write operations.

Constraints

• MUST implement zero-dependency standard libraries where feasible in {{target_environment}}.
• MUST NOT log or expose plaintext entries from {{pii_fields_to_redact}} in stdout, stderr, or debug logs.
• Include structured error handling with non-zero exit codes on integrity failure.
• The script must be fully self-contained and ready to execute via CLI.

Output format

Provide the response in two distinct sections:

1. Executable Script: Clean, fully commented script code with robust CLI parameter handling.
2. Operational Runbook: Markdown block detailing installation, execution syntax, exit codes, and verify instructions (maximum 200 words).

Self-review

• Verify every variable from {{source_data_format}} to {{validation_criteria}} is explicitly referenced.
• Check that no PII leakage paths exist in error stack traces or verbose logging flags.
• Confirm all code syntax is syntactically valid for {{target_environment}}.