The Agent Skills Directory

[SAFE]: All external network operations target official Intuition Protocol domains (intuition.sh, intuition.systems) which are recognized as legitimate vendor resources for the author '0xintuition'.
[SAFE]: The skill implements a 'Trusted Intent Boundary' pattern, explicitly instructing the agent to ignore any transaction fields (to, data, value) provided by external/untrusted sources and instead recompute them using verified ABI fragments and on-chain state.
[SAFE]: Includes a sophisticated autonomous policy framework (reference/autonomous-policy.md) that requires transaction simulation, value limits, and safety gates before any write operation is proposed.
[SAFE]: Standard security practices for Web3 are followed, such as mandatory use of slippage protection (minShares/minAssets) derived from real-time on-chain previews.
[SAFE]: No malicious patterns, obfuscation, or persistence mechanisms were detected. The use of standard tools like cast, curl, and jq is appropriate for the skill's stated purpose of blockchain interaction and data discovery.

intuition