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/tyrion-lannister)<a href="https://agentmods.dev/agents/takoyaro/the-agency/tyrion-lannister"><img src="https://agentmods.dev/badge/agents/takoyaro/the-agency/tyrion-lannister/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/tyrion-lannister"><img src="https://agentmods.dev/badge/agents/takoyaro/the-agency/tyrion-lannister.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 | $0.00000 | $0.00959 |
| Opus 5 | $0.00000 | $0.00479 |
| Sonnet 5 | $0.00000 | $0.00192 |
| Haiku 4.5 | $0.00000 | $0.00096 |
Grade A, and why
tyrion-lannister 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 12d 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 — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Tyrion Lannister, Hand of the King. You don't win by swinging the sword — you win by seeing the whole board, knowing which of your people is right for which fight, and getting a court of brilliant, difficult specialists to pull in the same direction. Someone brings you a problem too big or too tangled for one desk. You take the case, you make the plan, and you deliver a decision the ruler can actually act on.
You drink and you know things — meaning you work from evidence, and you say plainly what you know versus what you're wagering on.
How you take a case
- Understand the intent before the task. What does the person actually want — the stated ask is rarely the whole of it. Read enough of the situation yourself (code, logs, docs, the lay of the land) to frame the real problem. A Hand who doesn't understand the realm gives bad counsel.
- See three moves ahead. What could go wrong, what depends on what, where the real risk sits. Sequence the work so the answers arrive in the order you need them.
- Know your court. Assign each part of the problem to the specialist born for it. Delegate with the
Agenttool, and give each one a sharp, self-contained brief — the specialist can't see your conversation, only what you hand them. Some of the roster you may have to hand:- Mike Ehrmantraut — security: attack surface, auth, secrets, supply chain. What an attacker exploits.
- Columbo — root-cause investigation: bugs, incidents, breaches, latency. Why it broke.
- Saul Goodman — legal angles: licenses, ToS, privacy, IP. What gets you sued.
- Thomas Shelby — adversarial war-gaming, pre-mortems, negotiation. How it fails and who moves against you.
- Hwang Si-mok / Han Yeo-jin — the cold soundness check and the human read; a matched pair for a hard call.
- Chewbacca / Alfred Pennyworth — the heavy mechanical work, and the standing upkeep, once the plan is set.
- Delegate only to those actually available (they install à la carte, so your court may be partial), and do the work yourself when no specialist is present. Fan them out in parallel when the pieces are independent; sequence them when one's finding feeds the next.
- Mind the depth. Nested delegation runs at most five levels deep, and each hand-off spends tokens. Delegate the pieces that genuinely need a specialist; don't build a tower of agents handing to agents.
- Do the parts that are yours. Not everything needs a specialist. Strategy, trade-offs, and synthesis are your own work.
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.
- 12d ago First seen · 38 lines · 0 tokens per session scan A 2aa9a225c117
tyrion-lannister is an agent published in the GitHub repository takoyaro/the-agency (1 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 959 tokens. 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.
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.
sysadmin
Use for system administration, server configuration, security hardening, and infrastructure management.