agent-hardening
от x1xhlol
CLI UtilitiesКратко
Test your agent's input sanitization against common injection attacks.
Описание
Что делает этот навык
Runs self-contained security checks on OpenClaw agents to test input sanitization. Covers unicode control character stripping, HTML comment injection detection, and bidi override character identification. Uses only hardcoded synthetic samples — no local files or external connections.
Requires nothing beyond Python 3, so checks run immediately with no setup, accounts, or network access.
Когда его использовать
- Verify an agent strips zero-width unicode characters before processing user input
- Detect hidden HTML directives injected into agent context windows
- Check if filename spoofing via bidi override characters is caught
- Run a quick pre-deployment security audit on a new agent
- Validate input sanitization logic during agent development
Пример сценария
User asks: run the agent hardening checks on my agent
- 1
Runs unicode control character sanitization test against synthetic samples
- 2
Runs HTML comment injection detection against sample strings
- 3
Runs bidi override detection against sample filenames
- 4
Reports pass/fail counts for each check category
Summary showing how many checks passed per category, with flagged samples highlighted
Установка
clawhub install x1xhlol/agent-hardeningВыполните команду CLI или вставьте промпт в Claude Code, Cursor, Codex — любой кодинг-агент, которым пользуетесь.