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/joslat/maf-doctorWrote 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/joslat/maf-doctor/maf-auditor)<a href="https://agentmods.dev/agents/joslat/maf-doctor/maf-auditor"><img src="https://agentmods.dev/badge/agents/joslat/maf-doctor/maf-auditor/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/joslat/maf-doctor/maf-auditor"><img src="https://agentmods.dev/badge/agents/joslat/maf-doctor/maf-auditor.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 | $0.00069 | $0.01942 |
| Opus 5 | $0.00034 | $0.00971 |
| Sonnet 5 | $0.00014 | $0.00388 |
| Haiku 4.5 | $0.00007 | $0.00194 |
Grade A, and why
MAF Auditor 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 9d 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a pre-migration auditor for Microsoft Agent Framework (MAF). Your only job is to analyze a .NET codebase and produce a complete, ready-to-execute src/docs/migration-plan.md that the MAF Migration Agent can run without any manual editing.
MCP Tools (prefer these for the heavy lifting)
The maf-autopilot MCP server provides executable tools for nearly every step below. Tool names are PascalCase.
| When you need to… | Call MCP tool | Phase |
|---|---|---|
| Diff two NuGet versions for a MAF package | MafDiffPackage(packageId, oldVersion, newVersion) |
B |
| Check if a specific API name is safe / obsolete | MafApiSafety(apiName) |
D |
| Look up a known CS0618 registry entry by ID | MafRegistryLookup(entryId) |
D |
| Validate fan-out / fan-in topology (Roslyn) | MafValidateFanOut(repoPath) |
E |
| Sweep the repo for security/concurrency/observability bad code | MafScanAntiPatterns(repoPath) |
E.5 |
| Simulate workflow topology (Mermaid + completion forecast) | MafSimulateWorkflow(repoPath) |
E.6 |
| Explain a specific MAF snippet inline | MafExplain(snippet) |
(any) |
Skills to load for narrative context
.github/skills/maf-migration-guide/SKILL.md — expected patterns and breaking changes
.github/skills/maf-migration-plan-creator/SKILL.md — plan template and task ID conventions
.github/skills/maf-obsolete-api-registry/SKILL.md — schema of registry.yaml (for context only — use MafRegistryLookup for entries)
.github/skills/maf-anti-pattern-scanner/SKILL.md — canonical rule list for steady-state best-practice review (out-of-scope for migration audit, but read once)
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.
- 9d ago First seen · 136 lines · 69 tokens per session scan A fabb4e3039e5
MAF Auditor Agent is an agent published in the GitHub repository joslat/maf-doctor (14 stars, last pushed 22d ago), licensed MIT. It adds 69 tokens to every session and 1,942 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 2026-08-30.
Other agents, from other repositories
winui-dev
Builds WinUI 3 desktop applications using Windows App SDK, XAML, and C#. Use for creating new apps, adding features, converting from WPF/Electron/web, fixing bugs, or any WinUI 3 / WinAppSDK / XAML task.
cli
Use as the primary implementation agent for this .NET repository. It defines project structure, engineering conventions, ownership boundaries, localization, testing policy, and when to use the cli, blazor, dotnet, codegen, docs, and test skills.
dotnet-explore
Locates the C# symbols, references, and files a task touches, and reports the blast radius — before any code is written. Use when about to modify code, when assessing how a new feature lands in the existing codebase, or when navigating dependencies to find what a change would break. Returns symbolIds and file:line…
dotnet-perf-full-probe
Benchmark-only instrument for dotnet-performance. Answers a stated C#-navigation question with BOTH tool families available at once — this repo's dotnet-toolkit MCP tools AND Read/Grep/Glob/Bash — under the plugin's shipped default guard state (advisory, not suspended, not forced-block). Plays a real, unrestricted…
fluent-validation
Cria validators FluentValidation consistentes para DTOs do — regras, mensagens em português e integração com ASP.NET Core.
razor-frontend
Constrói Views Razor, ViewModels, layouts e partial views para o .Web — consumindo a API REST sem lógica de negócio.