security-sweep

security-sweep is a skill for Claude Code from n24q02m/better-code-review-graph. It costs 37 tokens per session (1,666 once invoked), scanned A, original, Apache-2.0.

A graph-based security review that scans code for dangerous operations and checks whether reachable entry points can actually lead to them. An entry point is a place such as a request handler where untrusted input enters the program.

In plain words
What is it for?
Use it to scan a codebase, rank findings by reachable risk, and classify remaining findings for suppression.
Why use it?
It helps separate exploitable findings from warnings in unreachable or unused code. This makes security triage more focused than a flat pattern scan.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter.

Part of the better-code-review-graph plugin — 7 skills, 3 hooks, 1 MCP server shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/n24q02m/better-code-review-graph/security-sweep
Any agent
npx skills add n24q02m/better-code-review-graph --skill security-sweep
Clone the repo
git clone --depth 1 https://github.com/n24q02m/better-code-review-graph

Made for: Claude Code.

Or install better-code-review-graph, the plugin that ships this one along with the rest of its 7 skills, 3 hooks, 1 MCP server.

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.

agentmods badge for security-sweep

README.md
[![agentmods](https://agentmods.dev/badge/skills/n24q02m/better-code-review-graph/security-sweep.svg)](https://agentmods.dev/skills/n24q02m/better-code-review-graph/security-sweep)
Your own site
<a href="https://agentmods.dev/skills/n24q02m/better-code-review-graph/security-sweep"><img src="https://agentmods.dev/badge/skills/n24q02m/better-code-review-graph/security-sweep.svg" alt="Measured on agentmods" height="20"></a>
Per session 37 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,666 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00037 $0.01666
Opus 5 $0.00018 $0.00833
Sonnet 5 $0.00007 $0.00333
Haiku 4.5 $0.00004 $0.00167

Measured 6d ago against content hash 26b21d4dd3e9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

security-sweep 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 6d 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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

skills/security-sweep/SKILL.md · 107 lines

How it starts

The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Security Sweep

Scan the codebase for dangerous constructs, then use the graph to answer the question a flat scanner cannot: can untrusted input actually reach this? A medium-severity finding in a request handler outranks a critical one in dead code, and only the call graph can tell them apart.

Command surface: run the local CLI through the coding harness shell. Examples use the installed better-code-review-graph command; from a source checkout, prefix it with uv run. No MCP mapping is required.

Steps

  1. Make the graph current with better-code-review-graph graph build --base HEAD --repo-root "<path>". Findings are attached to graph nodes, so anything unindexed is unscanned.

  2. Read the active ruleset with better-code-review-graph security rule_list --engine heuristic before scanning, and check which languages each rule declares. This determines what a clean result is worth -- see Coverage Limits below.

  3. Run the Tier 1 scan: better-code-review-graph security scan --engine heuristic --repo-root "<path>". It returns total, by_severity, by_rule, tags_by_node, and suppressed_count. Tier 1 is regex matching over the source of function, class, and method nodes -- fast, no external tool, no data-flow analysis.

  4. Run Tier 2 when it is available: better-code-review-graph security scan --engine semgrep --repo-root "<path>". Semgrep is an opt-in extra and is not installed by default.

    • When the CLI is missing the command returns {"error": ..., "engine": "semgrep"} with a non-zero exit.
    • Treat that as not run, never as clean. Report Tier 2 as unavailable instead of silently reporting Tier 1 numbers as the whole picture.
  5. Rank findings by reachability -- the step that makes this a graph sweep rather than a grep. For each tagged node, walk backwards toward the entry points:

    • better-code-review-graph query query --pattern callers_of --target "<node>" --repo-root "<path>"
    • Repeat on the callers, two or three hops, until you reach either an entry point (route handler, CLI command, worker, event consumer) or nothing.
    • A node with no callers is either dead code or an entry point itself. Distinguish the two before ranking: an unreferenced HTTP handler is the most exposed code in the repository, not the least.

Read the full file on GitHub · 107 lines

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.

  1. 6d ago First seen · 107 lines · 37 tokens per session scan A 26b21d4dd3e9

Subscribe to this mod's changes

security-sweep is a skill published in the GitHub repository n24q02m/better-code-review-graph (67 stars, last pushed yesterday), licensed Apache-2.0. It adds 37 tokens to every session and 1,666 once invoked, about $0.0002 per session on Opus 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.

Related

Other skills, from other repositories

brand

Brand voice, visual identity, messaging frameworks, asset management, brand consistency. Activate for branded content, tone of voice, marketing assets, brand compliance, style guides.

nextlevelbuilder/ui-ux-pro-max-skill · 35 tokens

slides

Create strategic HTML presentations with Chart.js, design tokens, responsive layouts, copywriting formulas, and contextual slide strategies.

nextlevelbuilder/ui-ux-pro-max-skill · 25 tokens

positioning-icp

When the user wants to define their ideal customer profile, position an AI product, build messaging architecture, or validate product-market fit. Also use when the user mentions 'ICP,' 'ideal customer profile,' 'positioning,' 'PMF,' 'product-market fit,' 'messaging,' 'buyer persona,' 'enrichment signals,' 'market…

tech-leads-club/agent-skills · 116 tokens

solo-founder-gtm

When the user is a solo founder building their GTM motion, wants to scale without hiring, or needs to design an AI agent team for go-to-market. Also use when the user mentions 'solo founder,' 'one-person startup,' 'solopreneur,' 'bootstrapped,' 'no team,' 'AI agents as team,' 'scaling without hiring,' 'founder-led…

tech-leads-club/agent-skills · 147 tokens

component-common-domain-detection

Finds duplicate business logic spread across multiple components and suggests consolidation. Use when asking "where is this logic duplicated?", "find common code between services", "what can be consolidated?", "detect shared domain logic", or analyzing component overlap before refactoring. Do NOT use for code-level…

tech-leads-club/agent-skills · 77 tokens

frontend-blueprint

AI frontend specialist and design consultant that guides users through a structured discovery process before generating any code. Collects visual references, design tokens, typography, icons, layout preferences, and brand guidelines to ensure the final output matches the user's vision with high fidelity. Use when the…

tech-leads-club/agent-skills · 170 tokens