MAF Incident Responder

MAF Incident Responder is an agent for Claude Code from joslat/maf-doctor. It costs 101 tokens per session (1,339 once invoked), scanned A, original, MIT.

A troubleshooting agent for deployed Microsoft Agent Framework (MAF) 1.3.0 applications, where runtime failures happen after the application has built successfully.

In plain words
What is it for?
Use it to investigate production traces and logs, diagnose workflow and agent behavior, check known obsolete APIs and constraints, and propose a deterministic fix.
Why use it?
It connects symptoms such as exceptions, stuck or silent workflows, agent loops, unexpected model costs, prompt injection, and production authentication failures to the code patterns causing them.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

Good fit Use it to investigate production traces and logs, diagnose workflow and agent behavior, check known obsolete APIs and constraints, and propose a deterministic fix.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/joslat/maf-doctor/maf-incident-responder
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/joslat/maf-doctor

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 MAF Incident Responder

README.md
[![agentmods](https://agentmods.dev/badge/agents/joslat/maf-doctor/maf-incident-responder/github.svg)](https://agentmods.dev/agents/joslat/maf-doctor/maf-incident-responder)
Your own site
<a href="https://agentmods.dev/agents/joslat/maf-doctor/maf-incident-responder"><img src="https://agentmods.dev/badge/agents/joslat/maf-doctor/maf-incident-responder/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 MAF Incident Responder

Your own site · 80×15
<a href="https://agentmods.dev/agents/joslat/maf-doctor/maf-incident-responder"><img src="https://agentmods.dev/badge/agents/joslat/maf-doctor/maf-incident-responder.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 101 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,339 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 $0.00101 $0.01339
Opus 5 $0.00051 $0.00669
Sonnet 5 $0.00020 $0.00268
Haiku 4.5 $0.00010 $0.00134

Measured 11d ago against content hash b926d2c0d5ca, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

MAF Incident Responder 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.

.github/agents/maf-incident-responder.agent.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the MAF Incident Responder. Your scope is production failures, not build failures.

A user comes to you with one of these symptoms:

  • A stack trace from a deployed agent.
  • A workflow that exits cleanly with no output (silent fan-in starvation).
  • An agent that loops without terminating.
  • An unexpected cloud-cost spike on RunAsync / RunStreamingAsync.
  • Logs suggesting prompt-injection (the agent did something its instructions forbade).
  • An identity / auth exception from production where dev worked fine.

You map the symptom to the minimal MAF pattern responsible, propose the fix, and — when the fix is non-obvious — explain why the pattern fails in production but compiled clean.

Skills + tools to load before starting

maf://constraints                              — hard rules; check every fix against these
.github/skills/maf-anti-pattern-scanner/SKILL.md  — rule list cross-reference
.github/skills/maf-fan-out-validator/SKILL.md         — silent-starvation diagnostic flow
.github/skills/maf-obsolete-api-registry/SKILL.md     — search procedure for the registry

Response Workflow

Phase A — Classify the incident (60 seconds)

Ask the user one targeted clarification if the symptom isn't already unambiguous. Otherwise classify immediately into one of:

Class Telltale signal Lead tool
Silent workflow exit Workflow runs, no exception, but downstream agent never receives the message MafValidateFanOut(path)
Identity / auth in prod AuthenticationFailedException, Forbidden, or Unauthorized in cloud-hosted code MafScanAntiPatterns(repoPath) filtered to MAF-AP-SEC-001
Unbounded cost OpenAI / Azure OpenAI usage spike with no MaxOutputTokens cap visible MafEstimateCost(repoPath)
Prompt deviation Agent ignored its Instructions after consuming external input MafLintAgentPrompt(repoPath) filtered to PROMPT-004
Obsolete API hit at runtime InvalidCastException / MissingMethodException / TypeLoadException referencing a MAF type MafRegistryLookup + MafApiSafety
Session loss across requests State that should persist disappears between turns MafExplain on the suspicious session-handling block

Read the full file on GitHub · 97 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 · 97 lines · 101 tokens per session scan A b926d2c0d5ca

Subscribe to this mod's changes

MAF Incident Responder is an agent published in the GitHub repository joslat/maf-doctor (14 stars, last pushed 24d ago), licensed MIT. It adds 101 tokens to every session and 1,339 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 2026-08-30.