Skill Claude CodeCodex
After a Codex hook block, do not reconstruct secrets. Silent auto-send is ingress Responses.
Non-Human Identity & Runtime Enforcement for AI Agents — open-core MCP/SDK to mask, enforce, and audit every agent call.
This repository also configures its own agents. See what blekline-oss tells them →
Skill Claude CodeCodex
After a Codex hook block, do not reconstruct secrets. Silent auto-send is ingress Responses.
Skill Claude CodeCodex
Install Blekline Codex hook wrappers (.sh and .cmd) via @blekline/codex-hooks init.
Skill Claude CodeCodex
Mask user text with PII or secrets before Codex tools or the model see it.
Skill Claude CodeCodex
Route downstream MCP tools through blekline-proxy for argument policy checks.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: