Codex
Prompts and configuration targeted at Codex (OpenAI’s hosted coding agent).
These pages stay noindex so leftover drafts cannot rank as
authoritative CVE floors. Humans browse the cards below; agents can
fetch the same bounded files through the recipe JSON feed and MCP
search tools.
What usually lives here
AGENTS.md— the repo brief Codex reads on every invocation. Covers how to build, how to test, style conventions, and what files are out-of-bounds.- Task prompts — the narrow description you paste into a Codex task that turns a finding into a PR.
PROJECT_GUIDELINESsnippets that are small enough to share but too big to inline in every task prompt.
Browse entries
Every entry carries its author, team, and maturity. Click any card for the full prompt.
Codex Vulnerable Dependency Remediation
developmentA Codex CLI prompt designed for non-interactive, CI-driven runs (codex exec --sandbox workspace-write --json). It picks up a single CVE / advisory id, locates the affected …
Codex Sensitive Data Remediation
developmentA Codex CLI prompt for headless SDE remediation — hard-coded secrets, PII in logs, credentials committed to source — wired for codex exec --sandbox workspace-write --json so it …