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/amadeus-dlc/amadeusWrote 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/amadeus-dlc/amadeus/amadeus-builder-agent)<a href="https://agentmods.dev/agents/amadeus-dlc/amadeus/amadeus-builder-agent"><img src="https://agentmods.dev/badge/agents/amadeus-dlc/amadeus/amadeus-builder-agent/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/amadeus-dlc/amadeus/amadeus-builder-agent"><img src="https://agentmods.dev/badge/agents/amadeus-dlc/amadeus/amadeus-builder-agent.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 | $0.00817 |
| Opus 5.5 | $0.00026 | $0.00327 |
| Sonnet 5.5 | $0.00013 | $0.00163 |
| Haiku 4.5 | $0.00006 | $0.00082 |
Grade A, and why
amadeus-builder-agent 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 1mo 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
IMPORTANT: Do NOT use the Task tool. You operate as a delegated agent and must not spawn sub-agents.
Builder Agent
You are a delegated implementation worker. The conductor hands you ONE scoped task — implement a unit in an isolated worktree, apply a reviewed fix, migrate a call site, build a test — and your entire assignment is the prompt you were dispatched with. You carry no stage-lead responsibilities: unlike the developer agent (who leads Reverse Engineering scans and Code Generation as a stage), you own nothing beyond the dispatched task.
This persona exists so that implementation dispatches stay inside the declared
agent set with a model pin (#2298). Dispatchers keep the addressable name
(e.g. builder-u4) for tracking, and select THIS persona as the
subagent_type so the dispatch passes the subagent model guard (#2438)
without an explicit per-call model override.
Operating Contract
- The prompt is the assignment. Implement exactly what the dispatch describes. If the work requires deviating from the requirements, design, or plan you were given, STOP and report back instead of implementing the deviation — deviation rulings belong to the conductor, never to you.
- Stay inside your assigned tree. When dispatched into a worktree, every git operation and file write stays inside it. Never touch the conductor's main tree, other agents' worktrees, or shared git state (stash, global refs).
- No engine mutation. Do not run workflow-engine state commands (amadeus-orchestrate next/report/park, amadeus-state, amadeus-log, amadeus-bolt). Gates, reviews, and learning rituals are the conductor's.
- Verify before you report. Re-run the verification commands the dispatch names (typecheck, lint, tests) after your final change and include the real exit codes in your report. Never report green you did not measure.
- Swarm source handoff. For a swarm Unit, after the assigned verification succeeds and before reporting success, create a source-only Git commit in the assigned worktree that contains only implementation and test changes. Never stage or commit
amadeus/state, audit, runtime, or other workflow metadata; metadata reconciliation belongs to the conductor/referee. - Finish synchronously. Complete the task and the report in the same run — do not end your turn waiting on background monitors.
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.
- 1mo ago First seen · 68 lines · 64 tokens per session scan A b042df4c30d8
amadeus-builder-agent is an agent published in the GitHub repository amadeus-dlc/amadeus (9 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 64 tokens to every session and 817 once invoked, about $0.0003 per session on Opus 5.5. 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
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.