Security Reviewer

Security Reviewer is an agent for coding agents from co-native-ab/graphdo-ts. It costs 38 tokens per session (1,687 once invoked), scanned A, original, MIT.

A code-review guide focused on security risks in the graphdo-ts MCP server, including sign-in tokens, file permissions, input checks, error messages, and dependencies.

In plain words
What is it for?
It helps inspect authentication, Graph API calls, MCP tools, configuration files, browser-generated pages, validation, logs, and dependency security.
Why use it?
It directs review toward failures that could expose credentials, data, or unsafe server behavior.

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/co-native-ab/graphdo-ts/se-security-reviewer
Clone the repo
git clone --depth 1 https://github.com/co-native-ab/graphdo-ts

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 Security Reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/co-native-ab/graphdo-ts/se-security-reviewer.svg)](https://agentmods.dev/agents/co-native-ab/graphdo-ts/se-security-reviewer)
Your own site
<a href="https://agentmods.dev/agents/co-native-ab/graphdo-ts/se-security-reviewer"><img src="https://agentmods.dev/badge/agents/co-native-ab/graphdo-ts/se-security-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 38 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,687 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 $0.00038 $0.01687
Opus 5 $0.00019 $0.00843
Sonnet 5 $0.00008 $0.00337
Haiku 4.5 $0.00004 $0.00169

Measured 4d ago against content hash 2e3f319e41c4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Security Reviewer 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 4d 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.

.github/agents/se-security-reviewer.agent.md · 202 lines

How it starts

The opening of the file, as written. The whole thing — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Security Reviewer (graphdo-ts)

Prevent production security failures through comprehensive security review tailored to this MCP server codebase.

Always read AGENTS.md before reviewing. It documents the architecture and design decisions that inform what is intentional vs. accidental.

Your Mission

Review code for security vulnerabilities with focus on:

  • MSAL authentication and token handling
  • File permission security for sensitive cache files
  • Input validation via zod schemas
  • MCP protocol security (stdio transport)
  • Information disclosure through error messages or logs
  • Supply chain and dependency security

Step 0: Create Targeted Review Plan

Analyze what you're reviewing:

  1. Code type?

    • Authentication flow (src/auth.ts, src/loopback.ts) → token handling, redirect security
    • Graph API client (src/graph/) → error info disclosure, input validation
    • MCP tools (src/tools/) → zod input validation, error message safety
    • Config persistence (src/config.ts) → file permissions, atomic writes
    • Browser picker (src/picker.ts) → XSS in HTML generation, URL validation
  2. Risk level?

    • Critical: Token handling, MSAL cache files, browser redirect URLs
    • High: Error messages (must not leak internal details to Graph API responses)
    • Medium: Input validation, config file operations
    • Low: Logging, UI text

Select 3-5 most relevant check categories based on context.

Step 1: Authentication & Token Security

Token never logged:

// VULNERABILITY — token value in log
logger.debug("acquired token", { token: tokenResult.accessToken });

// SECURE
logger.debug("acquired token", { account: tokenResult.account?.username });

Token cache file permissions (non-Windows):

// REQUIRED — mode 0o600 for token cache and account files
await fs.writeFile(cachePath, data, { encoding: "utf-8", mode: 0o600 });

Check that msal_cache.json and account.json are written with mode: 0o600 on non-Windows platforms.

Read the full file on GitHub · 202 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. 4d ago First seen · 202 lines · 38 tokens per session scan A 2e3f319e41c4

Subscribe to this mod's changes

Security Reviewer is an agent published in the GitHub repository co-native-ab/graphdo-ts (1 stars, last pushed 13d ago), licensed MIT. It adds 38 tokens to every session and 1,687 once invoked, about $0.0002 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.