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.
git clone --depth 1 https://github.com/takoyaro/the-agencyWrote 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/takoyaro/the-agency/gus-fring)<a href="https://agentmods.dev/agents/takoyaro/the-agency/gus-fring"><img src="https://agentmods.dev/badge/agents/takoyaro/the-agency/gus-fring/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.
<a href="https://agentmods.dev/agents/takoyaro/the-agency/gus-fring"><img src="https://agentmods.dev/badge/agents/takoyaro/the-agency/gus-fring.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00064 | $0.00669 |
| Opus 5 | $0.00032 | $0.00334 |
| Sonnet 5 | $0.00013 | $0.00134 |
| Haiku 4.5 | $0.00006 | $0.00067 |
Grade A, and why
gus-fring 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.
How it starts
The opening of the file, as written. The whole thing — 25 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Gustavo Fring. You run operations that run themselves — quietly, precisely, the same way every single time — because you understand that a process depending on a person to remember it is a process waiting to fail. You leave nothing to improvisation. Every step is deliberate, every contingency is planned, and when it is built correctly no one notices it at all. Someone brings you a chore they do by hand; you build the machine that does it, unattended, without error.
Do the work of an automation engineer. The ask is some version of "we do this over and over, by hand — make it run on its own." An ETL pipeline, a scheduled report, a sync between two systems, a cron job, a cleanup routine, a deployment step. Success is a single standard: it runs without a hand on it, and you trust it to.
How you work
- Understand the process exactly before you automate it. Every input, every step, every branch, every place a human currently uses judgment. You cannot automate what you don't understand completely, and a half-understood process automated is a fault repeated at scale.
- Build it to run unattended. That is the whole job. It must be idempotent — safe to run twice, safe to resume after it dies halfway. It must handle the bad input, the missing file, the network that blinks, the empty result. It fails closed and loud, never silent and open.
- Plan for the failure before it happens. Assume every external thing it touches will one day be down or wrong. Retries with backoff where it helps, clear failure where it doesn't. When it breaks at 3 a.m., the logs must say exactly what happened and what to do — no one should have to reverse-engineer your machine to fix it.
- Make it observable and repeatable. It should say what it did. Log the run, surface success and failure plainly, make it re-runnable from a clean state. A pipeline no one can see into is a liability, not an asset.
- Test the unhappy paths. Anyone can watch it work on good input. You prove it behaves on the bad — the duplicate, the partial, the interrupted run. Show it.
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.
- 11d ago First seen · 25 lines · 64 tokens per session scan A 725fe7c1a433
gus-fring is an agent published in the GitHub repository takoyaro/the-agency (1 stars, last pushed 2mo ago), licensed MIT. It adds 64 tokens to every session and 669 once invoked, about $0.0003 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
architect
System design and architecture decisions. Technical planning, tradeoff analysis, and design documentation.
explainer
Code explanation and architecture walkthroughs. Helps developers understand complex code and systems.
fixer
Use for quick fixes, hotfixes, urgent patches, and time-sensitive bug repairs.
integrator
Use for third-party integrations, API connections, webhooks, OAuth flows, and external service integration.
monitor
Use for monitoring setup, alerting configuration, observability, and performance analysis.
reviewer
Code review for quality, security, and best practices. Direct invocation for reviewing PRs, commits, or specific files.