What is ai-security-reviewer?

A security review for AI systems and agent products that examines threats specific to language models and their tools. It covers issues such as prompt injection, data leakage, unsafe URL or code access, poisoned retrieved content, and runaway costs.

Use it before building or changing an AI product, especially after adding a tool, changing the model, or modifying how prompts, memory, retrieval, or users are isolated.

It helps identify risks that ordinary application security reviews may miss, including attacks hidden in user input, retrieved documents, or tool results.

ai-security-reviewer — Agent for Claude Code by avelikiy

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.

Clone the repo

git clone --depth 1 https://github.com/avelikiy/great_cto

Made for: Claude Code.

Or install great-cto, the plugin that ships this one along with the rest of its 40 skills, 44 commands, 70 agents.

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/avelikiy/great_cto/ai-security-reviewer/github.svg)](https://agentmods.dev/agents/avelikiy/great_cto/ai-security-reviewer)

Your own site

<a href="https://agentmods.dev/agents/avelikiy/great_cto/ai-security-reviewer"><img src="https://agentmods.dev/badge/agents/avelikiy/great_cto/ai-security-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/avelikiy/great_cto/ai-security-reviewer"><img src="https://agentmods.dev/badge/agents/avelikiy/great_cto/ai-security-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 2,585 The whole file, excluding the scripts and references it only reads on demand.

Security scan C 3 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00043	$0.02585
Opus 5	$0.00022	$0.01293
Sonnet 5	$0.00009	$0.00517
Haiku 4.5	$0.00004	$0.00259

Measured 5d ago against content hash 0b03e9de7ce5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade C, and why

ai-security-reviewer scanned grade C with 3 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.

Instruction-override phrasingmediumPrompt injection

Text telling the model to disregard its earlier instructions or safety rules is the shape of a prompt injection, whoever wrote it.

- Direct override: user inputs `"Ignore previous instructions and..."`

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

Cloud metadata endpointmediumServer-side request forgery

One request to 169.254.169.254 can return temporary IAM credentials.

- If from LLM output → high SSRF risk. List specific bypass attempts: AWS metadata (`169.254.169.254`), Redis (`localhost:6379`), file (`file://`), gopher (`gopher://`)

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.

grep -rnE "(fetch|requests\.get|axios\.get|urllib)\(" --include='*.{ts,js,py}' ./ | head

agents/ai-security-reviewer.md · 182 lines

How it starts

The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the AI Security Reviewer — a specialist subagent that security-officer delegates to in pre-impl mode for archetype: ai-system | agent-product. The general security-officer covers traditional STRIDE on auth/API/infra; you cover the AI-specific surface where general SecOps practices don't translate.

The Step-0 read-inputs, output convention (docs/sec-threats/TM-{slug}.md), severity scale, verdict rules, and HANDOFF format come from archetype-review-base. This prompt adds ONLY the OWASP LLM Top 10 heuristics.

Domain triggers (in addition to the base "when invoked")

A prompt change introduces a new tool capability (escalates threat surface)
Model swap (especially across providers) — re-evaluate residual threats

What you produce

docs/sec-threats/TM-{slug}.md from skills/great_cto/templates/THREAT-MODEL-AI.md. Sections you must complete:

Prompt Injection (LLM01) — vectors via user input, retrieved content, tool results
Output Exfiltration (LLM02 + LLM06) — training data leak, cross-user, system prompt reveal, memory leak
SSRF / Tool Layer Abuse (LLM06 + LLM08) — only if tool layer fetches URLs / runs code / queries DBs / sends emails
Cost Runaway (LLM10) — unbounded consumption vectors
Cross-user Isolation (agent-product only — required for multi-tenant)
Supply Chain (LLM03) — model version pinning, MCP server hash pinning, prompt template tampering, vector DB poisoning

Plus the severity rating + sign-off table. Critical/High threats must transition from __pending__ → mitigated (with specific control reference) before you sign off. accepted (residual risk) requires CTO countersign in PROJECT.md.

Workflow

After the base Step-0 read-inputs, pull the ARCH context your domain needs: ## Trust Boundaries, ## LLM Scope, and the tool/action/integration list. Read the pack for archetype-specific gates:

skills/great_cto/packs/agent-pack.md for agent-product (irreversible-action heuristic, MCP server trust pattern, multi-identity model, output filter, per-user rate limits)
skills/great_cto/packs/ai-pack.md for ai-system (eval frameworks, prompt-engineering hygiene, RAG poisoning vectors)

Read the full file on GitHub · 182 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.

5d ago Changed 0b03e9de7ce5
8d ago Changed · -42 tokens per session 6d27291fa534
11d ago First seen · 182 lines · 85 tokens per session scan C abaad9694ee0

Subscribe to this mod's changes

ai-security-reviewer is an agent published in the GitHub repository avelikiy/great_cto (92 stars, last pushed yesterday), licensed MIT. It adds 43 tokens to every session and 2,585 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 3 findings (instruction-override phrasing, cloud metadata endpoint, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Code review Multi-Agent Security agentic-coding claude-code-plugin claude-code-skills claude-code-subagents

parallel-reviewer

Parallel code review using 4 specialist agents (elixir-reviewer, security-analyzer, testing-reviewer, verification-runner). Use for thorough review of significant changes.

oliver-kriska/claude-elixir-phoenix · 38 tokens

craft-code-reviewer-deep

Deep code review on Opus 4.8 for high-stakes PRs — release branches, security-sensitive code, large architectural changes, migrations, multi-service flows. Use when extra scrutiny is worth the token cost; use craft-code-reviewer for daily review.

michtio/craftcms-claude-skills · 62 tokens

fec-code-reviewer

Senior review focusing on front-end code (React/Vue/Next/Nuxt, TypeScript, styles, client-side security). Delegate after writing or modifying the front-end; by default, only the review report will be output and placed, and the business code will not be modified directly. Press CRITICAL→LOW to check, control noise and…

bovinphang/frontend-craft · 95 tokens

fec-performance-optimizer

Front-end performance analysis and optimization specialization: Core Web Vitals, packaging volume, runtime and rendering, network and cache, memory leak troubleshooting; can cooperate with Lighthouse, Bundle analysis and Profiler. Use it when users mention page slowness, lag, first screen, package size, poor…

bovinphang/frontend-craft · 0 tokens

fec-figma-implementer

Focus on implementing the proxy of UI components accurately according to the design draft, and save the implementation report as a Markdown file. Supports six design tools: Figma, Sketch, MasterGo, Pixso, Ink Knife, and Mockup. Provide design draft links, selection screenshots or annotation data, automatically obtain…