codex-cli-agent — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited codex-cli-agent (Agent Skill) and scored it 87/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 3 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 3 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Dispatches bounded tasks to the Codex CLI (codex binary). Uses the run_agent.py task router with cli=codex.
[!IMPORTANT] Default model: `gpt-5-codex`. RequiresOPENAI_API_KEYin environment. Thecodexbinary must be on PATH.
run_agent.pypython ./scripts/run_agent.py \
<PERSONA_FILE> <INPUT_FILE> <OUTPUT_FILE> "<INSTRUCTION>" \
--cli codex --model gpt-5-codexThe prompt is piped to codex exec via stdin — not passed as a positional arg — to avoid OS ARG_MAX limits and process listing exposure.
python ./scripts/run_agent.py \
agents/refactor-expert.md \
target.py \
refactor.md \
"Refactor this function and explain the top 3 changes." \
--cli codexpython ./scripts/run_agent.py \
agents/security-auditor.md \
target.py \
security.md \
"Find vulnerabilities. Rate severity: CRITICAL / MODERATE / MINOR." \
--cli codex --model gpt-4o| Use Case | Why codex |
|---|---|
| Code review / refactor | GPT-5 Codex is code-optimized |
| OpenAI endpoint routing | Use any OPENAI_BASE_URL-compatible target |
| Isolated sub-task to GPT-5 | No main agent context bleed |
agents/)| Persona | Use For |
|---|---|
security-auditor.md | Red team, vulnerability scanning |
refactor-expert.md | Code optimization, DRY, readability |
architect-review.md | System design, modularity |
echo "Say hello in one sentence." | codex exec --model gpt-5-codex -
python ./scripts/run_agent.py agents/refactor-expert.md target.py output.md "Summarize this file." --cli codexecho $OPENAI_API_KEY # must be set
which codex # must be on PATH
codex --version
codex exec --help # verify exec subcommand and available flags~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.