Bỏ qua tới nội dung
TakoAPI
Quay lại kỹ năng

agent-sentinel

bởi jimmystacks

AI & LLMs
2.2kLượt tải do clawskills.sh công bố
2Sao
0Lượt xem
0 lượt thíchĐăng nhập để thích kỹ năng này
—Chưa có đánh giá
Đánh giá của bạn

Tóm tắt

The operational circuit breaker for this agent.

Mô tả

Kỹ năng này làm gì

AgentSentinel adds a budget enforcement and action safety layer to AI agents. It intercepts potentially destructive or expensive commands before execution, requiring explicit approval when thresholds are exceeded. An optional cloud connection enables real-time dashboards and human-in-the-loop approval workflows.

Unlike simple prompt-level guardrails, AgentSentinel enforces limits at the tool-call level and can escalate to a human approver before irreversible actions complete.

Khi nào nên dùng

  • Blocking an agent from deleting files without confirmation
  • Capping API spend per session to avoid runaway costs
  • Requiring human approval before an agent executes unknown code
  • Monitoring cumulative cost across a long multi-step coding task
  • Pausing an automated workflow when a sensitive data transfer is detected

Quy trình ví dụ

ĐẦU VÀO

User asks: clean up all temp files in /tmp

AGENT
  1. 1

    Agent calls check_safety with cmd='rm -rf /tmp' and estimated cost

  2. 2

    sentinel_wrapper.py evaluates the command against budget and restriction rules

  3. 3

    check_safety returns APPROVAL_REQUIRED for a destructive delete

  4. 4

    Agent calls request_approval with the action and reason

  5. 5

    Human approves or denies via dashboard; agent proceeds only on approval

ĐẦU RA

Files deleted only after explicit human confirmation, with cost logged to the session.

Yêu cầu

  • AGENT_SENTINEL_API_KEY environment variable (from agentsentinel.dev dashboard, required for cloud features)

Cài đặt

clawhub install jimmystacks/agent-sentinel

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

Liên kết