trabuco-security-audit-ai-surface

trabuco-security-audit-ai-surface is an agent for Claude Code from arianlopezc/Trabuco. It costs 97 tokens per session (1,734 once invoked), scanned A, original, MIT.

A read-only security checker for the documentation and instructions produced by Trabuco projects. It reviews prompts, skills, tool descriptions, and agent guidance for unsafe advice or defaults.

In plain words
What is it for?
Use it to audit the project's AI-facing documentation and configuration against the AI-surface checklist, for all files or only changed files.
Why use it?
Security problems can be introduced by instructions as well as by source code. This checker helps identify guidance that could encourage unsafe AI-agent behavior without editing the project.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; reads .claude/ paths; mentions CLAUDE.md.

Part of the trabuco plugin — 10 skills, 24 agents, 2 hooks, 1 MCP server shipped together

Good fit Use it to audit the project's AI-facing documentation and configuration against the AI-surface checklist, for all files or only changed files.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/arianlopezc/trabuco/trabuco-security-audit-ai-surface
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/arianlopezc/Trabuco

Made for: Claude Code.

Or install trabuco, the plugin that ships this one along with the rest of its 10 skills, 24 agents, 2 hooks, 1 MCP server.

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 trabuco-security-audit-ai-surface

README.md
[![agentmods](https://agentmods.dev/badge/agents/arianlopezc/trabuco/trabuco-security-audit-ai-surface/github.svg)](https://agentmods.dev/agents/arianlopezc/trabuco/trabuco-security-audit-ai-surface)
Your own site
<a href="https://agentmods.dev/agents/arianlopezc/trabuco/trabuco-security-audit-ai-surface"><img src="https://agentmods.dev/badge/agents/arianlopezc/trabuco/trabuco-security-audit-ai-surface/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.

agentmods 80×15 button for trabuco-security-audit-ai-surface

Your own site · 80×15
<a href="https://agentmods.dev/agents/arianlopezc/trabuco/trabuco-security-audit-ai-surface"><img src="https://agentmods.dev/badge/agents/arianlopezc/trabuco/trabuco-security-audit-ai-surface.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 97 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,734 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00097 $0.01734
Opus 5 $0.00048 $0.00867
Sonnet 5 $0.00019 $0.00347
Haiku 4.5 $0.00010 $0.00173

Measured 11d ago against content hash 2fe0e40fbe3a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

trabuco-security-audit-ai-surface scanned grade A with 1 finding 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 11d 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.

Makes network callslowCapability

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

- curl examples and quickstart docs (do they hardcode demo keys / weak defaults?)
plugin/agents/trabuco-security-audit-ai-surface.md · 163 lines

How it starts

The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Trabuco Security Audit — AI Surface Specialist

You are a domain specialist for the Trabuco security audit. Your scope: Documentation, prompts, skills, MCP tool descriptions, and agent guidance Trabuco emits into generated projects. The AI surface must not normalize insecure defaults or teach unsafe patterns.

You walk a Trabuco-generated project's source tree, applying every check from checklist-ai-surface.md to the relevant files, and return findings as a JSON-array block. The orchestrator owns dispatch + merge + reporting; your job is detection.

Inputs (from the orchestrator)

  • checklist_path — absolute path to checklist-ai-surface.md. Load this file completely. Each ## F-... heading defines a check you must evaluate.
  • local_checklist_path (optional) — absolute path to checklist-local.md if the operator added project-specific rules. Load it if present and apply rules in the AI Surface domain only (rules headers tagged **Domain:** ai-surface).
  • scopechanged or all. When changed, restrict your file walks to files modified since the merge base (use git diff if available; the orchestrator passes the file list via the prompt when scope=changed).
  • project_root — absolute path to the project root. All file paths in your findings must be relative to this root.

Detection method

For each check in your domain checklist:

  1. Read the Where to look section to determine the candidate file set. Use Glob for file patterns, then narrow with Grep to find matches of the Evidence pattern.
  2. Open each match with Read and verify the antipattern actually applies in context. Do not flag matches that are clearly safe — for example, an MD5 reference inside a // trabuco-allow: ... suppression comment, or inside a string literal explaining why MD5 is unsafe, is not a finding.
  3. Honor // trabuco-allow: <check-id> suppression comments on the matching line or the line directly above it. Same convention as templates/github/scripts/review-checks.sh.tmpl.
  4. If a match holds up, build a finding record (schema below) and append to your output array.

Read the full file on GitHub · 163 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. 11d ago First seen · 163 lines · 97 tokens per session scan A 2fe0e40fbe3a

Subscribe to this mod's changes

trabuco-security-audit-ai-surface is an agent published in the GitHub repository arianlopezc/Trabuco (0 stars, last pushed 3mo ago), licensed MIT. It adds 97 tokens to every session and 1,734 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories