audit-file

audit-file is a command for coding agents from sembraniteam/claude-plugins. It costs 23 tokens per session (896 once invoked), scanned A, original, MIT.

A command for performing a detailed security review of one source file, with each finding linked to a CWE weakness category and supported by line-level evidence.

In plain words
What is it for?
Use it to inspect a file for issues such as unsafe database queries, shell commands, file paths, redirects, or access-control mistakes. CWE is a public list of common software weaknesses.
Why use it?
It focuses the review on one file and traces how input moves through the code, making security problems easier to verify and explain.

Command

Part of the security-auditor plugin — 1 skill, 7 commands, 3 agents 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 commands/sembraniteam/claude-plugins/audit-file
Clone the repo
git clone --depth 1 https://github.com/sembraniteam/claude-plugins

Or install security-auditor, the plugin that ships this one along with the rest of its 1 skill, 7 commands, 3 agents.

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 audit-file

README.md
[![agentmods](https://agentmods.dev/badge/commands/sembraniteam/claude-plugins/audit-file.svg)](https://agentmods.dev/commands/sembraniteam/claude-plugins/audit-file)
Your own site
<a href="https://agentmods.dev/commands/sembraniteam/claude-plugins/audit-file"><img src="https://agentmods.dev/badge/commands/sembraniteam/claude-plugins/audit-file.svg" alt="Measured on agentmods" height="20"></a>
Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 896 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 $0.00023 $0.00896
Opus 5 $0.00012 $0.00448
Sonnet 5 $0.00005 $0.00179
Haiku 4.5 $0.00002 $0.00090

Measured 3d ago against content hash 4fc2bb142664, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

audit-file 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 3d 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.

security-auditor/commands/audit-file.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are running a focused, line-level security audit on a single file using the security-auditor plugin.

The target file is: $ARGUMENTS

If no file path was provided, ask: "Which file should I audit? Please provide the path."

Step 1 — Read and understand the file

Read the file at the provided path. Identify:

  • File type (language, framework role: controller, model, service, config, etc.)
  • What kind of input does this file receive? (HTTP params, database results, file system, IPC, etc.)
  • What outputs or effects does it produce? (HTTP response, DB writes, file writes, shell execution, etc.)

Step 2 — Trace data flows

For each external input source, trace the data path through the file to all potential sinks:

  • Dangerous sinks: SQL query construction, shell execution, file path assembly, template rendering, eval/exec, deserialization, HTTP response body, redirect target
  • Intermediate transformations: note which ones sanitize/validate and which do not

Step 3 — Check each vulnerability category

Work through the OWASP Top 10 (A01–A10) + Extended Categories checklist from the secure-code-review skill, applied specifically to this file:

  1. Broken access control — does the file enforce authorization before sensitive operations? IDOR, CORS misconfiguration, SSRF (outbound HTTP calls with user-controlled URLs)
  2. Security misconfiguration — debug flags, verbose error messages, permissive CORS, missing security headers
  3. Software supply chain failures — any import/require/use of third-party libraries? Flag for CVE check; unsigned/unverified build artifacts
  4. Cryptographic failures — hash algorithms, encryption schemes, IV/salt handling
  5. Injection (SQL, NoSQL, OS, LDAP, template) — trace every DB call and exec
  6. Insecure design — missing rate limiting, trust boundary violations
  7. Authentication failures — session handling, token validation, password checks
  8. Software or data integrity failures — insecure deserialization (pickle, yaml.load, JSON.parse with __proto__), unsigned updates/plugins
  9. Security logging & alerting failures — missing audit trail for sensitive ops, PII in logs, log injection
  10. Mishandling of exceptional conditions — verbose error messages leaking internals, fail-open logic, unhandled exceptions

Read the full file on GitHub · 72 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. 3d ago First seen · 72 lines · 23 tokens per session scan A 4fc2bb142664

Subscribe to this mod's changes

audit-file is a command published in the GitHub repository sembraniteam/claude-plugins (2 stars, last pushed 1mo ago), licensed MIT. It adds 23 tokens to every session and 896 once invoked, about $0.0001 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.