What is security-review?

A six-phase security review workflow for checking code changes against the main branch, the primary shared version of the code.

It reviews command-line input, secret handling, daemon communication, software dependencies, and how the code trusts AI-generated content.

It organizes different security checks and removes duplicate findings when a review is run again.

How sageox/ox configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to sageox/ox. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/sageox/ox/main/.claude/skills/security-review/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/sageox/ox

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/sageox/ox/security-review/github.svg)](https://agentmods.dev/skills/sageox/ox/security-review)

Your own site

<a href="https://agentmods.dev/skills/sageox/ox/security-review"><img src="https://agentmods.dev/badge/skills/sageox/ox/security-review/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/sageox/ox/security-review"><img src="https://agentmods.dev/badge/skills/sageox/ox/security-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 151 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,771 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector warn 7 Sept 2026

SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

medium Excessive Agency · line 75
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00151	$0.01771
Opus 5.5	$0.00060	$0.00708
Sonnet 5.5	$0.00030	$0.00354
Haiku 4.5	$0.00015	$0.00177

Measured 12d ago against content hash 8bb0f6f64d64, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

security-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/security-review/SKILL.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/security-review — ox AI security pipeline

You are orchestrating a Synthesia-style 6-phase security review over the user's diff against origin/main. The pipeline shape, the dedup-before-validate ordering, and the right-size-models-per-phase principle all come from that post; the ox specifics (threat model, CLI/daemon primitives, hunter perspective frames) are local.

Trigger phrases

/security-review (no args) — review the diff vs origin/main. Default.
/security-review --scope=<path-glob> — narrow to a specific path.
/security-review --hunter=<name> — run only one hunter (debug). Valid names: cli-input, secrets-redaction, daemon-ipc, supply-chain, llm-trust.
/security-review --cap=<usd> — raise the per-run cost cap (default $8; persisted in security/config.yml).

What you do

You are not the pipeline. You are the dispatcher. You shell out to security/scripts/orchestrate.sh and surface its output to the user concisely. The pipeline runs the AI subagents itself; do not try to re-implement them in this skill body.

bash security/scripts/orchestrate.sh "$@"

The orchestrator drives all six phases:

Prep — compute scope (diff vs origin/main), write security/.output/scope.md, and build the review input: the diff itself, test files excluded, split into chunks under diff.chunk_bytes (security/.output/review/).
Map — run security/scripts/deterministic.sh (parallel OSS scanners; each reports ran / skipped / failed) + one Cartographer call (Haiku) per chunk, reading the diff, to map entry points (CLI commands, daemon IPC handlers) to sinks. Writes security/.output/surface.md.
Hunt — per chunk, 5 hunter subagents in parallel (Sonnet), each reading the diff, the surface map and scope. Each has an explicit perspective frame (cli-input / secrets-redaction / daemon-ipc / supply-chain / llm-trust) to fight finding convergence. Writes security/.output/findings-raw.jsonl.
Dedup — single Sonnet pass merges hunter findings + deterministic findings by root cause. Writes security/.output/findings-deduped.jsonl.
Validate — one call per finding, model split: Sonnet for ~90%, Opus for findings in the hard classes (secrets-redaction, daemon-ipc, supply-chain; security/config.yml hard_classes). Stricter than hunters; traces real call paths; checks existing mitigations.
Aggregate — apply the coverage gate, drop false-positives, rank by severity, emit security/.output/FINDINGS.md (markdown) + security/.output/findings.sarif (machine).

Read the full file on GitHub · 82 lines

Files

What ships with it

18 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

12d ago Changed 8bb0f6f64d64
18d ago Changed ba4a12e40c37
19d ago Changed · +5 lines 9ba124f6c151
1mo ago First seen · 77 lines · 151 tokens per session scan A edd33acf24ba

Subscribe to this mod's changes

security-review is a skill published in the GitHub repository sageox/ox (65 stars, last pushed today), licensed MIT. It adds 151 tokens to every session and 1,771 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Claude Code Code review Security agentic-engineering codex-cli droid gemini-cli hivemind

plan-eng-review

Engineering plan review. Reviews a plan document for architecture, code quality, test coverage, and performance before implementation begins. Produces a coverage diagram, failure modes map, and parallelization strategy. Use when asked to "review this plan", "eng review", or "architecture review".

HECer/yoke · 61 tokens

review

Pre-merge code review — the single canonical review of a change before it lands. Covers BOTH diff safety/structure (SQL safety, LLM trust-boundary violations, conditional side effects) AND engineering quality (architecture fit, edge cases, test coverage, performance). Use when asked to "review this PR", "code review"…

HECer/yoke · 103 tokens

audit

Use when the user asks to audit this repo, run the engineering audit, run a practice audit, or check the codebase against the engineering-audit rules pack. Drives the engineering-audit MCP server's tools through a full audit run and produces a self-contained HTML report.

rodlunt/engineering-audit · 57 tokens

ox: Skill for Claude Code