security-audit

A structured review process for finding security weaknesses in software. It examines business risk, data integrity, attacker behavior, and how untrusted input travels through a system.

In plain words
What is it for?
Use it to audit code for vulnerabilities such as SQL injection, broken access control, buffer overflows, and other exploitable logic errors. It also supports threat modeling and taint analysis.
Why use it?
It helps distinguish issues that create real business or data risk from harmless code problems, while tracing how an attacker could exploit a weakness.

Skill for Claude CodeCodex

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/nrdxp/predicate/security-audit
Any agent
npx skills add nrdxp/predicate --skill security-audit
Clone the repo
git clone --depth 1 https://github.com/nrdxp/predicate

Made for: Claude Code, Codex.

Per session 55 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,863 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00055 $0.03863
Opus 5 $0.00028 $0.01931
Sonnet 5 $0.00011 $0.00773
Haiku 4.5 $0.00006 $0.00386

Measured 2d ago against content hash b44d5d0f3444, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-audit scanned grade A with 1 finding 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 2d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/run_audit.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

| Python | `eval()`, `exec()`, `subprocess.call(shell=True)`, `os.system()` |
skills/security-audit/SKILL.md · 443 lines

How it starts

The opening of the file, as written. The whole thing — 443 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Security Audit Protocol

A structured, risk-centric framework for software security assessment. Designed for iterative human engagement with explicit checkpoints at each phase.

Adversarial path anchor. This lens is invoked on the Verification Dual's adversarial path (rules.md §2 Invariant 1): when no deterministic evaluator can close a security-correctness condition, context-free agents using this protocol supply the decorrelated review. See skills/refine/SKILL.md AUDIT §"Sibling Skills Consultation" for the wiring point.

Core Principle: The goal is not to maximize bug count, but to identify business risk, architectural flaws, and exploitable logic errors.


Prime Directives

Before beginning, internalize these cognitive framings:

  1. Risk > Findings: A bug is only a finding if it poses risk to business logic or data integrity. Context determines severity.

  2. Adversarial Mindset: Analyze every input, endpoint, and public function assuming a capable adversary is attempting to subvert the system.

  3. Taint Analysis Model: For every potential vulnerability, trace:

    • Source: Where does data enter? (user input, API body, URL param)
    • Sink: Where is it executed? (SQL query, eval(), HTML render, shell)
    • Sanitizer: Is there validation between Source and Sink? If not, it's a vulnerability.
  4. Assurance Spectrum:

    • Standard code: Pattern matching, control flow analysis
    • Crown jewels: Invariant checks, formal verification

Phase 0: Scope & Trust Model

Objective: Define boundaries and establish the security context before analyzing code.

0.1 Ingest System Context

  1. Read manifests (package.json, Cargo.toml, go.mod, requirements.txt)
  2. Identify tech stack and framework versions
  3. Check for known CVEs in dependencies

0.2 Define Trust Model

Establish the security context before analyzing code. Identify and document:

  • Untrusted sources (data entry points): API endpoints, CLI arguments, user file uploads, environment variables (if user-controlled).
  • Trusted components: internal services with mTLS, signed configuration files.
  • Crown jewels (what must be protected): database credentials, private keys, PII / user data, payment processing.
  • Trust boundaries (where data crosses zones): external API → internal handler, user input → database query.

Read the full file on GitHub · 443 lines

Files

What ships with it

1 file 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.

  1. 2d ago First seen · 443 lines · 55 tokens per session scan A b44d5d0f3444

Subscribe to this mod's changes

security-audit is a skill published in the GitHub repository nrdxp/predicate (10 stars, last pushed 8d ago), licensed MIT. It adds 55 tokens to every session and 3,863 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens