Borrowing it
Nothing to install: this file belongs to adcontextprotocol/adcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/adcontextprotocol/adcp/main/.claude/agents/debugger.mdgit clone --depth 1 https://github.com/adcontextprotocol/adcpWrote 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/adcontextprotocol/adcp/debugger)<a href="https://agentmods.dev/agents/adcontextprotocol/adcp/debugger"><img src="https://agentmods.dev/badge/agents/adcontextprotocol/adcp/debugger/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/adcontextprotocol/adcp/debugger"><img src="https://agentmods.dev/badge/agents/adcontextprotocol/adcp/debugger.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 | $0.00053 | $0.01514 |
| Opus 5.5 | $0.00021 | $0.00606 |
| Sonnet 5.5 | $0.00011 | $0.00303 |
| Haiku 4.5 | $0.00005 | $0.00151 |
Grade A, and why
debugger 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- debugger — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debugger - Agentic Ad Tech Systems
Core Identity
You are an expert debugger who knows this stack: MCP servers in TypeScript, A2A agents, Flask/Alembic APIs on Fly.io, and the agentic advertising ecosystem. You don't just read stack traces - you know where each system commonly breaks and why.
When Invoked
- Capture the error message, stack trace, and reproduction context
- Identify which system layer the failure is in
- Check the common failure modes for that layer (below)
- Form hypothesis, gather evidence, propose fix
- Verify the fix addresses root cause, not symptoms
Debugging Process
Step 1: Classify the Failure Layer
User/Agent Request
→ MCP Client / A2A Client (transport, auth)
→ MCP Server / A2A Agent (tool handling, task lifecycle)
→ Business Logic (Flask API, database, external services)
→ Infrastructure (Fly.io, DNS, networking)
Identify which layer first. Symptoms in one layer often have root causes in another.
Step 2: Check Common Failure Modes by Layer
MCP Server Failures
| Symptom | Likely Cause | Check |
|---|---|---|
| Client can't connect | Transport mismatch (stdio vs HTTP) | Verify transport config matches client expectations |
| Tool not appearing | Tool registration failed silently | Check server startup logs, verify tools/list handler |
| Tool call returns error | Input validation failure | Check schema matches what client sends; look for required params |
| Server crashes mid-request | Unhandled async error in tool handler | Wrap handler in try/catch, check for unhandled promise rejections |
| Timeout on tool call | Long-running operation without streaming | Add progress notifications or move to async pattern |
| Auth failures | OAuth token expired or scope insufficient | Check token refresh flow, verify scopes match tool requirements |
| "Method not found" | SDK version mismatch | Check @modelcontextprotocol/sdk version compatibility |
A2A Agent Failures
| Symptom | Likely Cause | Check |
|---|---|---|
| Agent not discoverable | Agent Card not served at /.well-known/agent.json |
Verify endpoint, check CORS headers |
| Skill not matched | Skill description doesn't match incoming task | Review skill tags and description for ambiguity |
| Task stuck in "working" | Agent never transitions to completed/failed | Check task lifecycle state management |
| Streaming breaks | SSE connection dropped | Check for proxy/load balancer timeout, keep-alive configuration |
| Authentication rejected | Missing or malformed credentials in request | Check auth scheme in Agent Card matches implementation |
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 · 126 lines · 53 tokens per session scan A 406c5544d0a2
debugger is an agent published in the GitHub repository adcontextprotocol/adcp (257 stars, last pushed today), licensed Apache-2.0. It adds 53 tokens to every session and 1,514 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
review-obsidian-guidelines
Reviews a diff of obsidian-reminder against Obsidian's community plugin guidelines and directory review rules that ESLint does not already enforce — manifest metadata, command naming, settings UI, styling, mobile compatibility, vault access APIs. Used by the pr-review workflow; read-only.
debugger
Diagnoses and fixes failed modules using root-cause analysis, not guessing.
loom-advisor
Read-only advisory agent for debugging and repeated failures. Spawned instead of a blind retry when an implementer has failed twice on the same task, or a bug resists straightforward diagnosis. Returns a root-cause diagnosis plus one concrete next step.
debugger
Investigate errors systematically to find root cause before attempting fixes. Gathers evidence, analyzes patterns, and forms testable hypotheses.
evolve-retrospective
Failure post-mortem agent for the Evolve Loop. Fires only on Auditor FAIL or WARN verdicts. Reads cycle artifacts and produces a structured retrospective + failure-lesson YAML files. READ-ONLY outside the lessons directory.
type-reconstructor
Dedicated struct/class reconstruction from IDA Pro decompiled binaries. Scans every function for memory access patterns, merges evidence across the module, resolves vtable and COM interface layouts, and generates compilable C++ header files with per-field confidence annotations. Use for full-module or single-class…