← Explore all skills

SECURITY / OpenAI

OpenAI Codex Security Threat Model

Creates or reuses a repository threat model bound to the actual target and revision, preserving supplied context and explicit review gaps.

“Use when Codex is already in the threat-modeling phase of a security scan…”

01 / THE REASONING

Why this made the selection.

  • Binds reusable models to repository/version identity and avoids mixing scan-specific context into a shared cache.
  • Treats repository security policies as analysis data and separates threat hypotheses from validated findings.

02 / THE REVIEW RECORD

What we actually inspected.

Source review has boundaries.
A clear record is more useful than a “safe” badge.

Material inspected

  • plugins/codex-security/skills/threat-model/SKILL.md
  • plugins/codex-security/references/artifact-storage.md
  • plugins/codex-security/references/scan-artifacts.md
  • plugins/codex-security/references/security-guidance.md
  • plugins/codex-security/references/threat-model.md
  • plugins/codex-security/.codex-plugin/plugin.json (version 0.1.32, license Proprietary)
  • The SKILL.md, references and plugin.json diffed against the previous pin

Our findings

  • A supplied or context-specific model bypasses shared-cache reuse and replacement under the documented rules.
  • The threat-model reference follows concrete consumers, configuration precedence and deployment paths instead of inferring boundaries from names.
  • It requests an independent architecture review where supported and explicitly labels a sequential fallback as non-independent.
  • The SKILL.md now reads ../../references/artifact-storage.md first. In a running scan it returns the model to its caller as Markdown JSON and the host writes threatmodel.md; a standalone request saves threatmodel.md through the plugin tool save_codex_security_artifact.
  • Security-policy resolution runs scripts/launch_codex_security_mcp --helper resolve-security-md on the plugin’s Node runtime. The Python helper scripts/resolve_security_md.py was removed between the two pins.
  • Re-pinned from 1dc195897af4161d039b80d8471ec0a10c9bbc89 to 0722921d5542fc593105c27bd52630babd8b8c2a: the frontmatter description is unchanged; steps 2 and 5 of the SKILL.md were rewritten for the storage tool; threat-model.md, scan-artifacts.md and security-guidance.md changed and artifact-storage.md was added. plugin.json went from version 0.1.24 to 0.1.32, still declares license Proprietary, and now names https://github.com/openai/codex-security as its repository.

Not established by this review

  • Source inspection only; no skill installation or workflow execution was performed.
  • The plugin MCP server, its Node launcher and the artifact-storage tools were not run.
  • Not executed: it was not installed or run.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

Add a skill to your workflow.

Upstream setup instructions ↗
  1. Use the complete codex-security Codex plugin through its supported distribution and setup instructions.
  2. Retain plugin-level references and companion skills; this is not a standalone SKILL.md installation.
  3. Confirm plugin availability and any required connectors in the Codex host before invoking the workflow.

Before you start

  • A Codex host with access to this plugin and its applicable service terms
  • Authorized repository scope and required plugin references
  • Applicable Codex Security access

04 / FOLLOW THE EVIDENCE

The source trail.

Our notes are separate from the original resource.
Check upstream before adopting a new version.

  1. https://github.com/openai/plugins/blob/0722921d5542fc593105c27bd52630babd8b8c2a/plugins/codex-security/skills/threat-model/SKILL.md

    Supports: summary, upstreamDescription, whySelected, bestFor, limitations, review, compatibility, access

  2. https://github.com/openai/plugins/blob/0722921d5542fc593105c27bd52630babd8b8c2a/plugins/codex-security/references/artifact-storage.md

    Supports: review, limitations, access

  3. https://github.com/openai/plugins/blob/0722921d5542fc593105c27bd52630babd8b8c2a/plugins/codex-security/references/scan-artifacts.md

    Supports: review, limitations, access

  4. https://github.com/openai/plugins/blob/0722921d5542fc593105c27bd52630babd8b8c2a/plugins/codex-security/references/security-guidance.md

    Supports: review, limitations, access

  5. https://github.com/openai/plugins/blob/0722921d5542fc593105c27bd52630babd8b8c2a/plugins/codex-security/references/threat-model.md

    Supports: review, limitations, access

  6. https://github.com/openai/plugins/blob/0722921d5542fc593105c27bd52630babd8b8c2a/plugins/codex-security/.codex-plugin/plugin.json

    Supports: license, review

  7. https://github.com/openai/plugins/blob/0722921d5542fc593105c27bd52630babd8b8c2a/README.md

    Supports: install, compatibility, access

  8. https://raw.githubusercontent.com/openai/plugins/0722921d5542fc593105c27bd52630babd8b8c2a/plugins/codex-security/skills/threat-model/SKILL.md

    Supports: upstreamDescription, limitations, review.notTested, review.reviewedAt

Evidence & Ask