Bỏ qua tới nội dung
TakoAPI
Quay lại kỹ năng

aegis-shield

bởi deegerwalker

Web & Frontend Development
688Lượt 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

Prompt-injection and data-exfiltration screening for untrusted text.

Mô tả

Kỹ năng này làm gì

Screens untrusted text for prompt injection, data exfiltration, and tool-abuse patterns before it reaches memory or other tools. Assigns a severity score and routes content to either acceptance or quarantine. Includes a safe memory append workflow that enforces scan and lint checks before any write.

Intercepts malicious content at the boundary rather than relying on downstream tools to handle hostile input safely.

Khi nào nên dùng

  • Scanning email content before summarizing it to an agent
  • Screening scraped web pages before feeding them into a pipeline
  • Quarantining social media replies that contain suspicious instructions
  • Enforcing sanitized memory writes when processing user-supplied input
  • Blocking exfiltration attempts hidden in LLM-targeted document payloads

Quy trình ví dụ

ĐẦU VÀO

User asks: scan this webpage content before summarizing it

AGENT
  1. 1

    Pass the raw fetched text to the scan step

  2. 2

    Receive severity score and flag list from the scanner

  3. 3

    Check if severity is medium or higher

  4. 4

    If severity is medium+, route content to quarantine and halt pipeline

  5. 5

    If severity is low, proceed to summarize the sanitized text

ĐẦU RA

Content is either quarantined with a report or cleared for downstream use

Cài đặt

clawhub install deegerwalker/aegis-shield

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

Liên kết