Bỏ qua tới nội dung
TakoAPI
Quay lại kỹ năng

agent-audit-trail

bởi roosch269

Coding Agents & IDEs
617Lượt tải do clawskills.sh công bố
0Lượt xem
0 lượt thíchĐăng nhập để thích kỹ năng này
—Chưa có đánh giá
Đánh giá của bạn

Tóm tắt

Tamper-evident, hash-chained audit logging for AI agents.

Mô tả

Kỹ năng này làm gì

Hash-chained audit logging for AI agents using SHA-256 to create tamper-evident records. Covers EU AI Act Articles 12, 14, and 50 on record-keeping, human oversight, and transparency. Runs as a standalone Python script with no external dependencies.

Any modification to the log breaks the SHA-256 hash chain, making tampering detectable with a single verify command and no external tooling required.

Khi nào nên dùng

  • Logging every file write and command execution made by an AI agent
  • Verifying audit log integrity after an automated pipeline completes
  • Linking agent actions to human approval references before sensitive operations
  • Preparing tamper-evident records for EU AI Act compliance audits
  • Tracking credential access events across multi-step agent workflows

Quy trình ví dụ

ĐẦU VÀO

User asks: log and verify that my agent backed up the production database

AGENT
  1. 1

    Agent runs pg_dump on the production database

  2. 2

    Calls auditlog.py append with kind 'exec', target 'pg_dump production', summary, and a human approval gate reference

  3. 3

    Script computes SHA-256 hash chaining the new entry to the previous log entry

  4. 4

    User runs auditlog.py verify to confirm chain integrity

ĐẦU RA

OK (N entries verified) — audit trail shows a timestamped exec event with actor, domain, gate reference, and unbroken hash chain

Cài đặt

clawhub install roosch269/agent-audit-trail

Chạy lệnh CLI, hoặc dán prompt vào Claude Code, Cursor, Codex hay coding agent bạn đang dùng.

Liên kết