메뉴
Router that maps user security intent to specific files, scripts, and workflows inside the codex-security plugin. Guides any agent to find and invoke the right plugin files for security tasks — scans, threat models, validation, attack paths, triage, and tracking.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Reference and apply the Joint Strike Fighter (JSF) Air Vehicle C++ Coding Standards (Doc. No. 2RDU00001 Rev C, "AV Rules") — the Lockheed Martin/MISRA-derived safety-critical C++ standard used in avionics and other high-assurance software. Use this skill whenever the user asks to review, audit, or lint C++ code against the JSF/AV rules or against safety-critical C++ coding standards generally; asks about a specific "AV Rule" by number; wants to write safety-critical, MISRA-style, or DO-178B-adjacent C++ code; asks about coding standards for embedded, avionics, automotive, or other high-integrity C++ systems; or references "JSF", "Air Vehicle coding standards", "2RDU00001", or similar. Also use it to explain the rationale behind a specific rule, find which rule(s) a code snippet violates, or draft a project's own coding standard based on this one.
Apply systems-thinking to reinterpret fictional narratives (films, series, games). Deconstructs official explanations by identifying logical/thermodynamic inconsistencies, then proposes a coherent alternative that reframes all plot elements through a single system-level mechanism. Use when: analyzing sci-fi lore, debunking popular fan theories, reinterpreting fictional worldbuilding, or exploring "what if the official story is wrong." Strong triggers: "film theory", "lore reinterpretation", "debunk this theory", "systems thinking for fiction", "what are [X] actually for in [fictional system]", "reanalyze this story", "plot hole as feature". Anti-triggers: factual documentary analysis, real-world engineering, news reporting.
Use this skill before merging, praising, or considering "done" ANY new algorithm, heuristic, scoring function, or statistical technique added to a research-driven codebase — anything with a formula in it. Trigger this whenever a change is justified by theoretical reasoning alone ("this fills a gap," "this is more rigorous," "this generalizes X") rather than a measured before/after comparison. Also trigger when reviewing a pull request or diff that adds a new scoring/weighting term, when a component has never been run end-to-end against a real target, or when asked "does this actually help" about any subsystem. This skill exists because many projects repeatedly ship mathematically-broken code that passed code review and passing tests, because correctness-in-isolation was mistaken for correctness-in-practice.
Extracts the reasoning behind a solved problem — not just the diff or the fix, but why that approach was chosen, what alternatives were rejected and why, and what generalizes. Use this any time a debugging session, code review, research task, proof attempt, or exploratory investigation reaches a resolution and the user wants it captured as a learnings note, not just closed out. Trigger on explicit requests ("write up what we learned", "extract the approach", "capture this as a pattern", "document the reasoning", "turn this into a note") AND proactively suggest it whenever a nontrivial fix or finding is about to be left undocumented — a solution without a learnings note is unfinished work.
Full Python project bootstrapping workflow. Use this skill whenever the user wants to build a new Python tool, library, CLI, or module from scratch — especially when they mention "create X", "build X in Python", "write a Python project for X", or ask for a proper project with tests, linting, versioning, or git setup. Triggers on any request to scaffold, initialize, or structure a new Python project. Even if the user only says "build me X in Python", apply this skill — it encodes the full professional workflow: SPEC → implementation → pytest → README → lint → git. Always use this skill rather than improvising a one-off script when the deliverable is a reusable project.
Implement and run Inverse Rubric Optimization (IRO) experiments: a black-box judge testbed where an agent must reverse-engineer hidden evaluation criteria under a label budget. Use this skill whenever the user wants to: benchmark agent science methodologies, measure how well an optimizer recovers a hidden rubric, build a poetry/text IRO harness, study reward hacking in LLM optimization loops, measure performance vs label-budget tradeoffs, implement the normalized gap metric (blind vs rubric-visible baseline), or replicate/extend the Fulcrum IRO testbed. Also trigger for: "agent optimizer loop", "black-box judge hill-climbing", "label budget experiment", "prompt optimization against hidden criteria", "reward hacking in LLM eval loops", "IRO testbed", or any task involving an LLM iteratively optimizing a generation policy against an opaque scoring function.
| name | codex-security-adapter |
| description | Router that maps user security intent to specific files, scripts, and workflows inside the codex-security plugin. Guides any agent to find and invoke the right plugin files for security tasks — scans, threat models, validation, attack paths, triage, and tracking. |
| tags | ["security","scanning","threat-model","vulnerability"] |
Caveat: while the Codex-security plugin is proprietary, it is open source and anyone can download it and use it. In this adapter we provide the reference for any agentic harness to be able to use it outside Codex. You may still need to install it through official channels.
Always maintain TODOs.
Guide any agent to find and invoke the right codex-security plugin files for security tasks. This skill is a router — it maps user intent to specific files, scripts, and workflows inside ~/.codex/.tmp/plugins/plugins/codex-security/.
All paths below are relative to:
~/.codex/.tmp/plugins/plugins/codex-security/
Canonical variable: PLUGIN=~/.codex/.tmp/plugins/plugins/codex-security
→ $PLUGIN/skills/security-scan/SKILL.md — orchestrator skill, runs 4 phases in order:
Supporting references:
$PLUGIN/skills/security-scan/references/scan-artifacts-and-ledger.md$PLUGIN/skills/security-scan/references/repository-wide-scan.md$PLUGIN/skills/security-scan/references/repo-wide-artifacts-and-ledger.md$PLUGIN/skills/security-scan/references/repo-wide-high-impact-families.md$PLUGIN/skills/security-scan/references/repo-wide-instance-expansion.md$PLUGIN/skills/security-scan/references/repo-wide-validation-closure.md→ $PLUGIN/skills/security-diff-scan/SKILL.md — Git-backed change set review
Supporting: $PLUGIN/scripts/generate_rank_input.py — generates worklists from diffs
→ $PLUGIN/skills/deep-security-scan/SKILL.md — multi-pass, worker-specific threat models
→ $PLUGIN/skills/threat-model/SKILL.md
→ $PLUGIN/skills/threat-model/references/threat-model-guidance.md
→ $PLUGIN/skills/finding-discovery/SKILL.md — candidate finding discovery
→ $PLUGIN/skills/validation/SKILL.md
→ $PLUGIN/skills/validation/references/validation-guidance.md
→ $PLUGIN/skills/attack-path-analysis/SKILL.md
→ $PLUGIN/skills/attack-path-analysis/references/attack-path-facts.md
→ $PLUGIN/skills/attack-path-analysis/references/severity-policy.md
→ $PLUGIN/skills/fix-finding/SKILL.md
→ $PLUGIN/skills/triage-finding/SKILL.md
→ $PLUGIN/skills/triage-finding/references/triage-result-contract.md
→ $PLUGIN/skills/triage-finding/references/github-rest-intake.md
→ $PLUGIN/skills/track-findings/SKILL.md
→ $PLUGIN/skills/track-findings/references/jira.md
→ $PLUGIN/skills/track-findings/references/github-security-advisories.md
These scripts work without the MCP server. Run them directly.
# From a diff (PR/commit/branch)
python $PLUGIN/scripts/generate_rank_input.py make-diff-rank-input \
--repo <repo_root> --base <base_ref> --mode revisions --head <head_ref> \
--out <output_dir>/rank_input.jsonl
# From a local patch
python $PLUGIN/scripts/generate_rank_input.py make-diff-repo-input \
--repo <repo_root> --base <base_ref> --mode local-patch \
--out <output_dir>/rank_input.jsonl
# Copy to deep review input
python $PLUGIN/scripts/generate_rank_input.py copy-deep-review-input \
--rank-input <output_dir>/rank_input.jsonl \
--out <output_dir>/deep_review_input.jsonl
python $PLUGIN/scripts/finalize_scan_contract.py \
--scan-dir <scan_dir> --source-root <repo_root>
python $PLUGIN/scripts/validate_report_format.py \
--scan-dir <scan_dir>
python $PLUGIN/scripts/config_preflight.py \
--profile <security_scan|security_diff_scan|deep_security_scan> \
--plugin-dir $PLUGIN
# Full state management — call with subcommands:
# create-workspace, save-workspace, start-scan, complete-scan,
# fail-scan, get-scan, update-progress, list-findings, export-findings, etc.
python $PLUGIN/scripts/workbench_db.py <subcommand> [args...]
python $PLUGIN/scripts/validate_tracking_source.py # validate sealed scan for tracking
python $PLUGIN/scripts/snapshot_sqlite.py # SQLite snapshot
python $PLUGIN/scripts/filesystem_identity.py # filesystem identity helpers
python $PLUGIN/scripts/finding_preview.py # bounded finding details
python $PLUGIN/scripts/report_projection.py # JSON → markdown report
python $PLUGIN/scripts/workbench_source_excerpt.py # source excerpts from sealed revisions
python $PLUGIN/scripts/config_preflight.py # capability preflight check
Use these to validate or generate scan artifacts:
| Schema | Path | Purpose |
|---|---|---|
| Scan manifest | $PLUGIN/schemas/scan-manifest.schema.json | Top-level completed scan document |
| Findings | $PLUGIN/schemas/findings.schema.json | Array of finding objects with severity, CWE, attack paths |
| Coverage | $PLUGIN/schemas/coverage.schema.json | Scan completeness, surfaces, dispositions |
$PLUGIN/examples/completed-scan/scan-manifest.json$PLUGIN/examples/completed-scan/findings.json$PLUGIN/examples/completed-scan/coverage.json| File | When to read |
|---|---|
$PLUGIN/references/scan-artifacts.md | Any scan — defines artifact directory layout |
$PLUGIN/references/final-report.md | Final reporting — JSON authoring + markdown projection |
$PLUGIN/references/scan-contract.md | Machine-readable scan contract semantics |
$PLUGIN/references/finding-detail-fields.md | Rich finding fields (codeEvidence, rootCause, attackPath) |
$PLUGIN/references/sarif-adapter.md | SARIF 2.1.0 export |
$PLUGIN/references/static-finding-assessment.md | Static evidence assessment guidance |
$PLUGIN/references/config-preflight.md | Preflight workflow and remediation |
$PLUGIN/references/shared-hard-rules.md | Hard rules for every scan workflow |
For agents that cannot use the MCP server (non-Codex environments):
security-scan/SKILL.md)The SKILL.md files reference each other via $threat-model, $finding-discovery, $validation, $attack-path-analysis — treat these as "load and execute that skill's workflow."
Default scan output goes to:
<tmpdir>/codex-security-scans-<random>/
<scan_id>/
scan-manifest.json
findings.json
coverage.json
threat-model.md
discovery/
rank_input.jsonl
deep_review_input.jsonl
worklist.jsonl
coverage-ledger.jsonl
validation/
per-finding/
<finding-id>/
validation-report.md
attack-path/
per-finding/
<finding-id>/
attack-path-report.md
Override with --scan-dir on Python script calls.