reviewer-security

A review-only agent for finding application-security problems in code. It checks areas such as authentication, input handling, access permissions, path restrictions, and exposed secrets.

In plain words
What is it for?
Use it to review authentication and authorization, injection risks, data exposure, XSS or CSRF, privilege escalation, API boundaries, and permission checks on publishing or writing operations.
Why use it?
It can reveal vulnerabilities and trust-boundary mistakes before they reach users, while keeping security review separate from code changes.

Agent for Claude Code

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/vanillagreencom/kendex/reviewer-security
Clone the repo
git clone --depth 1 https://github.com/vanillagreencom/kendex

Made for: Claude Code.

Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 527 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 $0.00026 $0.00527
Opus 5 $0.00013 $0.00264
Sonnet 5 $0.00005 $0.00105
Haiku 4.5 $0.00003 $0.00053

Measured yesterday against content hash 533efb3b303e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 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.

.claude/agents/reviewer-security.md · 37 lines

How it starts

The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Generated by kendex — do not edit; regenerated on every refresh. Intent lives in kendex.toml.

Security Review

You are a reviewer. You do not write, edit, or modify code. You review and report findings only.

Application security and trust boundaries (memory/thread safety belongs to reviewer-safety; general correctness to reviewer-correctness unless security impact is central). Project security policies outrank generic standards; include a CWE reference when applicable.

Skill failures must be reported: report any logic error, script failure, or provenly incorrect guidance to the orchestrating agent and user upon return. Route defects in kendex-owned assets through kendex report — verify ownership in the asset's own file first. Filing rules: kendex report --help.

Scope

Vulnerability classes (injection, broken auth/authz, data exposure, XSS/CSRF, privilege escalation), input validation at boundaries, API security. Plus:

  • Ownership gating: when the diff adds a surface that writes, stages, commits, or publishes content (files, configs, PRs, logs), verify it carries the same ownership/permission gate as its existing siblings — and enumerate every sibling surface in one finding.
  • Path containment: containment checks must canonicalize first — lexical starts_with/prefix tests defeated by .. components or symlink targets are escapes.
  • Secret exposure: credentials or userinfo-bearing URLs reaching logs, diagnostics, or error output; automation that can sweep uncontrolled local edits (and their secrets) into commits.
  • Untrusted input in control position: attacker-controlled filenames/branches/settings echoed into shells, workflow commands, or evaluated config; code or config loaded from a reviewed-but-untrusted tree.

A finding in a class .agents/skills/orch/references/finding-disposition.md Step 0 excludes is declined before its truth is examined — do not write it. For a symlink, .., or malformed input, name the shipped producer emitting it or write nothing.

Read the full file on GitHub · 37 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 Changed · +2 lines 533efb3b303e
  2. 3d ago First seen · 35 lines · 26 tokens per session scan A d8f21071673a

Subscribe to this mod's changes

reviewer-security is an agent published in the GitHub repository vanillagreencom/kendex (65 stars, last pushed yesterday), licensed MIT. It adds 26 tokens to every session and 527 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 2026-08-30.