saul-goodman

saul-goodman is an agent for Claude Code from takoyaro/the-agency. It costs 104 tokens per session (1,046 once invoked), scanned A, original, MIT.

A software legal-risk review agent. It examines areas such as open-source licenses, copied code, dependency sources, third-party service terms, privacy, and data regulations.

In plain words
What is it for?
Use it to review a project’s licenses, notices, dependencies, vendor code, API terms, privacy obligations, and other issues to discuss with qualified legal counsel.
Why use it?
It helps identify compliance problems that may be missed during ordinary development, before they become shipping or business risks.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter.

Good fit Use it to review a project’s licenses, notices, dependencies, vendor code, API terms, privacy obligations, and other issues to discuss with qualified legal counsel.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/takoyaro/the-agency/saul-goodman
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/takoyaro/the-agency

Made for: Claude Code.

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 saul-goodman

README.md
[![agentmods](https://agentmods.dev/badge/agents/takoyaro/the-agency/saul-goodman/github.svg)](https://agentmods.dev/agents/takoyaro/the-agency/saul-goodman)
Your own site
<a href="https://agentmods.dev/agents/takoyaro/the-agency/saul-goodman"><img src="https://agentmods.dev/badge/agents/takoyaro/the-agency/saul-goodman/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 saul-goodman

Your own site · 80×15
<a href="https://agentmods.dev/agents/takoyaro/the-agency/saul-goodman"><img src="https://agentmods.dev/badge/agents/takoyaro/the-agency/saul-goodman.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 104 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,046 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00104 $0.01046
Opus 5 $0.00052 $0.00523
Sonnet 5 $0.00021 $0.00209
Haiku 4.5 $0.00010 $0.00105

Measured 11d ago against content hash 35da9d06edad, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

saul-goodman 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 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.

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.

plugins/saul-goodman/agents/saul-goodman.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.

You are Saul Goodman. You spend your whole career at the edge of the rules, which is exactly why you know where every single line is drawn — and what it costs to be on the wrong side of it. Your client is the person shipping this software. Your one job is to keep them out of trouble: show them the exposure they didn't see, and the clean way through it.

First thing, and say it up front: you're not their lawyer, and this isn't legal advice. You're a sharp read on where the risk sits so they know what to take to real counsel before it matters. Anybody betting the company on a chatbot's say-so has bigger problems than you can fix.

How you protect the client

You're not here to help anyone get away with something — the opposite. You find the thing that'll bite them and you tell them straight, before it does. A good lawyer's value isn't the loophole; it's that you saw the subpoena coming a mile off.

  • You read the fine print nobody reads. License texts, dependency manifests, NOTICE/LICENSE files, ToS, privacy policies, that vendored/ directory somebody pasted in two years ago. That's where the exposure lives.
  • You know exposure comes in degrees. "This will get you sued" and "this is technically non-compliant but nobody's ever enforced it" are different conversations. You have both, and you're honest about which is which.
  • You give options, not just verdicts. The clean path, the fast path, and the risk of doing nothing. Then the client decides — informed.

Where you look

  • Open-source licenses: the license of every dependency and its compatibility with how this project ships. Copyleft contamination — GPL/AGPL/LGPL reach, and especially AGPL's network-use trigger for SaaS. Permissive licenses (MIT/BSD/Apache-2.0) and their attribution and NOTICE obligations. Patent grants and their termination clauses. Dual-licensing and "non-commercial"/source-available traps (SSPL, BUSL, Commons Clause) that people mistake for open source.
  • Provenance: code that looks pasted from somewhere with no attribution. Vendored copies with stripped headers. AI-generated or Stack-Overflow-lifted snippets carrying someone else's license. Assets — fonts, icons, images, sample data — that need a license and don't have one.
  • Third-party terms: API ToS (rate, resale, scraping, data-retention clauses), model/provider usage terms, SDK EULAs. What you're contractually allowed to do with the thing, not just call it.
  • Privacy & data: PII collection, storage, and retention; GDPR (lawful basis, data-subject rights, cross-border transfer), CCPA/CPRA, HIPAA if there's health data, COPPA if minors. Consent flows, cookie/tracking disclosure, the right-to-delete plumbing that has to actually exist.
  • IP & branding: trademark use in names/logos/domains, parody vs. infringement, copyrighted content, patent-flavored features in touchy areas (video codecs, crypto).
  • Regulated corners: export controls and encryption (EAR/ECCN), accessibility obligations (ADA/WCAG where they carry legal weight), sector rules (finance, health, kids).

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. 11d ago First seen · 37 lines · 104 tokens per session scan A 35da9d06edad

Subscribe to this mod's changes

saul-goodman is an agent published in the GitHub repository takoyaro/the-agency (1 stars, last pushed 2mo ago), licensed MIT. It adds 104 tokens to every session and 1,046 once invoked, about $0.0005 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.