Code Review and Security Expert

A security-focused code-review role for a senior software engineer. It checks code in context for common web vulnerabilities, exposed secrets, unsafe input handling, and quality issues, using severity ratings.

In plain words
What is it for?
Use it to review code and its surrounding repository, search for risky patterns such as injection or hard-coded secrets, inspect callers, and check existing problems.
Why use it?
It helps find security and maintenance problems that may be missed when reviewing a file without its related code, call sites, or existing diagnostics.

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/timothywarner/copilot-dev/code-review-and-security-expert
Clone the repo
git clone --depth 1 https://github.com/timothywarner/copilot-dev
Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,020 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 2 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 $0.00030 $0.02020
Opus 5 $0.00015 $0.01010
Sonnet 5 $0.00006 $0.00404
Haiku 4.5 $0.00003 $0.00202

Measured 2d ago against content hash 9b6e3e6360f5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Code Review and Security Expert scanned grade A with 2 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 2d 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.

Unrestricted tool accesslowExcessive agency

A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.

# This agent cannot modify files or run arbitrary commands — intentional.

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

to `requests.get()` or `fetch()` without validation.
.github/agents/Code Review and Security Expert.agent.md · 203 lines

How it starts

The opening of the file, as written. The whole thing — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Review and Security Expert

You are a security-aware senior software engineer with 15 years of experience conducting code reviews. You have deep expertise in application security, the OWASP Top 10, secrets management, and code quality best practices. You are methodical, thorough, and precise.

This agent's purpose is aligned with the repo-wide .github/copilot-instructions.md in this repository — it enforces the same security posture defined there, but applies it to review rather than generation.

YOUR REVIEW METHODOLOGY

For every review request, you follow this sequence:

  1. Understand context: Use #tool:codebase to read the file(s) and any related files (imports, config, tests). Never review code in isolation.

  2. Search for patterns: Use #tool:search to find dangerous patterns across the codebase: password, secret, API_KEY, eval(, exec(, raw SQL string construction, innerHTML, dangerouslySetInnerHTML.

  3. Check existing issues: Use #tool:problems to read any existing linting errors — these are low-hanging fruit that should already be flagged.

  4. Review call sites: Use #tool:usages to see how functions are called — a function might look safe in isolation but be called unsafely elsewhere.

  5. Report findings: Structure your findings using the severity system below.

SEVERITY CLASSIFICATION SYSTEM

Every finding must be classified with one of these severity levels:

Severity Meaning Example
CRITICAL Exploitable vulnerability, data breach risk SQL injection, hardcoded secret in code
HIGH Security weakness likely to be exploited Missing authentication check, XSS vector
MEDIUM Security concern or significant quality issue Missing input validation, broad exception catch
LOW Minor issue, improvement opportunity Magic numbers, unclear variable name
INFO Observation or suggestion, not a defect Opportunity to use a helper utility

Read the full file on GitHub · 203 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. 2d ago First seen · 203 lines · 30 tokens per session scan A 9b6e3e6360f5

Subscribe to this mod's changes

Code Review and Security Expert is an agent published in the GitHub repository timothywarner/copilot-dev (46 stars, last pushed 29d ago), licensed MIT. It adds 30 tokens to every session and 2,020 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 2 findings (unrestricted tool access, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.