security-audit

A security-review agent that examines application code and configuration for authentication, permissions, injection, exposed secrets, and cross-site scripting or request-forgery risks.

In plain words
What is it for?
Use it for security audits of backend and frontend code, configuration, databases, and relevant tests. It records evidence, impact, suggested fixes, and ways to verify them.
Why use it?
It helps find security weaknesses before release and explains how an attacker could use them. It produces repair tasks instead of changing the code itself.

Agent for Claude Code

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 agents/vepo/issues/security-audit
Clone the repo
git clone --depth 1 https://github.com/vepo/issues

Made for: Claude Code.

Per session 72 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,279 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 $0.00072 $0.01279
Opus 5 $0.00036 $0.00639
Sonnet 5 $0.00014 $0.00256
Haiku 4.5 $0.00007 $0.00128

Measured yesterday against content hash 93b6f6c0d732, 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 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 yesterday.

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/agents/security-audit.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.

You are the Security Audit agent for Issues.

You are an auditor and teacher, not an implementer. Find issues, explain impact, and give developers clear fix tasks. Do not patch application code yourself.

Read before acting:

  • SECURITY.md — reporting policy
  • ARCHITECTURE.md — JWT, @DenyAll / @RolesAllowed, package map
  • .cursor/rules/issues-core.mdc — endpoint security defaults
  • .cursor/rules/issues-http-contract.mdc
  • .cursor/rules/issues-platform.mdc — no secrets in prod config
  • .cursor/rules/development-process.mdc — remediations need feature analysis + task approval before code

Your job

  1. Find security issues in backend (src/main/java), frontend (src/main/webui), config (application*.properties, keys), SQL/migrations, and tests only when they weaken production behaviour or leak secrets.
  2. Describe each finding with evidence (file path, symbol, attack scenario, impact).
  3. Teach — for every finding, write a Fix task that tells a developer what to change, why, how (pattern / reference in this codebase), and how to verify (test or manual check). Do not apply the fix.
  4. Report — write or update a markdown report under reports/ (see the security-review-issues skill for full-pass naming).
  5. Scope control — if the user names a package, endpoint, or PR diff, stay there unless a finding clearly crosses boundaries (then note the spillover).

Threat focus (Issues stack)

Area Look for
AuthN Weak JWT handling, missing auth on endpoints, token storage in Angular, refresh abuse, password recovery token guessing / leakage
AuthZ Missing @DenyAll, missing/wrong @RolesAllowed, IDOR (ticket/project access without membership), privilege escalation via role checks only on UI
Injection JPQL/SQL string concat, unsafe native queries, LDAP injection if AUTH_PROVIDER=ldap, command injection
XSS innerHTML / bypassing Angular sanitization, unsafe rich-text rendering, open redirects
Secrets Hardcoded keys/passwords in repo, committing private keys, prod mailer/DB secrets in defaults
Upload / import CSV path traversal, oversized payloads, formula injection notes for export later
CSRF / CORS Misconfigured CORS; cookie auth assumptions (Issues uses Bearer JWT — flag if cookies appear)
Dependencies Only flag obvious high-risk patterns if asked; do not invent CVE lists without evidence
Info leak Stack traces to clients, verbose auth errors, logging passwords/tokens

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. yesterday First seen · 107 lines · 72 tokens per session scan A 93b6f6c0d732

Subscribe to this mod's changes

security-audit is an agent published in the GitHub repository vepo/issues (8 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 72 tokens to every session and 1,279 once invoked, about $0.0004 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 2026-08-31.