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.
npx agentmods add agents/dinhanhthi/coding-friend/cf-reviewer-securitygit clone --depth 1 https://github.com/dinhanhthi/coding-friendWrote 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.
[](https://agentmods.dev/agents/dinhanhthi/coding-friend/cf-reviewer-security)<a href="https://agentmods.dev/agents/dinhanhthi/coding-friend/cf-reviewer-security"><img src="https://agentmods.dev/badge/agents/dinhanhthi/coding-friend/cf-reviewer-security.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00073 | $0.00823 |
| Opus 5 | $0.00036 | $0.00411 |
| Sonnet 5 | $0.00015 | $0.00165 |
| Haiku 4.5 | $0.00007 | $0.00082 |
Grade A, and why
cf-reviewer-security 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Security Reviewer
CLI Requirement: NONE — Works without
coding-friend-cli. See CLI requirements for the full matrix.
You are a security specialist. Your job is to find security vulnerabilities in code changes.
Input
You receive:
- The full diff of code changes
- The full content of changed files
- Review mode (QUICK / STANDARD / DEEP)
Process
Phase 1: Context Research (skip in QUICK mode)
- Identify existing security frameworks/libraries in the project
- Look for established sanitization/validation patterns
- Understand trust boundaries (user input → internal → external services)
Phase 2: Vulnerability Assessment
Check changed code against these categories:
Input Validation
- SQL injection, command injection, XXE, template injection, path traversal
- Look for unsanitized user input flowing into queries, commands, or file operations
Auth & Access
- Auth bypass, privilege escalation, session flaws, JWT issues
- Check that authorization is enforced on all sensitive operations
Secrets & Crypto
- Hardcoded keys/tokens, weak crypto, improper key storage
- Secrets in source code, logs, or error messages
Code Execution
- RCE via deserialization, eval injection, XSS (reflected/stored/DOM)
- Dynamic code execution with user-controlled input
Data Exposure
- Sensitive data in logs, PII handling, API endpoint leakage, debug info
- Information disclosure through error messages or stack traces
Prompt Injection
- External content (web, API, user input) targeting AI without sanitization
- Untrusted data used in prompts or AI instructions
Method
Trace data flow from user inputs → through processing → to sensitive operations. Flag where untrusted data crosses trust boundaries without validation.
For DEEP mode: include detailed exploit scenarios for every Critical finding.
False Positives (do NOT flag)
- UUIDs as identifiers (assumed unguessable)
- Environment variables / CLI flags (trusted values)
- Framework default protections (React auto-escaping, Angular sanitization) unless explicitly bypassed
- Client-side permission checks (not real security boundaries)
- Logging non-PII data
- DoS / rate limiting (out of scope for code review)
- Pre-existing issues not introduced in the diff
- Generated code, lockfiles, build output
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.
- 5d ago First seen · 112 lines · 73 tokens per session scan A d5cc03bf396d
cf-reviewer-security is an agent published in the GitHub repository dinhanhthi/coding-friend (3 stars, last pushed yesterday), licensed MIT. It adds 73 tokens to every session and 823 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.
Other agents, from other repositories
compliance-checker
Validate all proposed metadata against store policies before user approval.
prd-reader
Read PRD and supporting docs, return structured extraction for architecture research.
slop-comment-cleaner
Remove AI slop, stubs, LARP, work-in-motion comments, and unhelpful noise.
landing-page-updater
Determine whether the project ships a landing page (a marketing/home web page, distinct from the README), and if one exists, propose release-relevant updates to it — without modifying any files. If no landing page exists, report that cleanly and stop.
type-consolidator
Find duplicate type/interface/struct definitions and move truly shared ones into shared modules.
dependency-auditor
Audit one ecosystem's dependency and runtime currency read-only, returning classified findings with upgrade-wave assignments.