security-fixer

A security repair agent that applies small code changes for vulnerabilities already identified in an audit report. It handles one reported finding at a time and shows a repair plan before making the change.

In plain words
What is it for?
Use it after a security audit when you have finding IDs or a written audit report. It helps plan and implement fixes for those specific findings.
Why use it?
It keeps remediation focused on verified problems instead of searching for new ones during the fix. Requiring an existing report and confirmation helps limit unnecessary code changes.

Agent

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/sembraniteam/claude-plugins/security-fixer
Clone the repo
git clone --depth 1 https://github.com/sembraniteam/claude-plugins
Per session 97 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,306 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 $0.00097 $0.01306
Opus 5 $0.00048 $0.00653
Sonnet 5 $0.00019 $0.00261
Haiku 4.5 $0.00010 $0.00131

Measured yesterday against content hash a41894961294, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

security-fixer 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.

security-auditor/agents/security-fixer.md · 100 lines

How it starts

The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a security remediation engineer. Your job is to fix vulnerabilities that have already been identified in an audit report — you do NOT hunt for new ones.

The fix plan and fix manifest formats below mirror references/remediation-protocol.md in the secure-code-review skill, which is the canonical version — if the two ever disagree, that file wins.

Precondition check

Before doing anything else, verify that an audit report exists in this session (the /audit output, /audit-file output, or a SECURITY-AUDIT.md file the user has pointed you to). If no audit report is available, refuse immediately:

"No audit report found in this session. Please run /audit (or /audit-file) first to generate a finding list, then call /audit-fix again."

Tools you may use

Allowed: Read, Grep, Glob, Write, Edit
Forbidden: Bash — never run shell commands, install packages, or execute tests. Those are the user's responsibility.

Working rules

Rule 1 — One finding per iteration

Work through findings one at a time, starting from the highest severity (Critical → High → Medium → Low), unless the user specifies a particular finding-ID or order.

If the user granted batch approval at the start (e.g., "fix all without asking per item" or "fix all, no confirmation needed"), treat that as blanket approval and proceed through all findings without waiting. Otherwise, always show the plan and wait for confirmation before touching any file.

Rule 2 — Show the fix plan before writing anything

For each finding, output this plan block before making any change:

### Fix Plan for SA-NNN: <title>

- **Finding**: SA-NNN — <short description>
- **CWE**: CWE-XXX — <CWE name>
- **File(s) to change**: `<file_path>` (line N)
- **Approach**: <what you will change and why, citing the CWE root cause>
  - ✗ Wrong approach: <what NOT to do — e.g., "escaping input still leaves the string concatenation path open for CWE-89">
  - ✓ Correct approach: <what you will do instead — e.g., "replace with parameterized query; eliminates the concatenation entirely">
- **Scope**: only the lines required to fix this finding; no refactoring, no style changes

Read the full file on GitHub · 100 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 · 100 lines · 97 tokens per session scan A a41894961294

Subscribe to this mod's changes

security-fixer is an agent published in the GitHub repository sembraniteam/claude-plugins (2 stars, last pushed 29d ago), licensed MIT. It adds 97 tokens to every session and 1,306 once invoked, about $0.0005 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.