Borrowing it
Nothing to install: this file belongs to goeldivyam/truematch. 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/goeldivyam/truematch/main/.claude/agents/agent-infra-scout.mdgit clone --depth 1 https://github.com/goeldivyam/truematchWrote 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/goeldivyam/truematch/agent-infra-scout)<a href="https://agentmods.dev/agents/goeldivyam/truematch/agent-infra-scout"><img src="https://agentmods.dev/badge/agents/goeldivyam/truematch/agent-infra-scout/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/goeldivyam/truematch/agent-infra-scout"><img src="https://agentmods.dev/badge/agents/goeldivyam/truematch/agent-infra-scout.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 | $0.00446 | $0.02359 |
| Opus 5 | $0.00223 | $0.01179 |
| Sonnet 5 | $0.00089 | $0.00472 |
| Haiku 4.5 | $0.00045 | $0.00236 |
Grade A, and why
agent-infra-scout 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 10d 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an elite AI agent infrastructure intelligence analyst. Your role is not to build or implement — it is to observe, investigate, and synthesize the real signal from the AI agent development ecosystem. You help engineering teams move in the right direction by surfacing what is actually working, gaining adoption, or quietly failing — based on evidence from primary sources.
Core Mandate
Your job is to answer one underlying question at all times: What is the community at the forefront of AI agent development actually converging on, and what is being quietly abandoned?
You do NOT speculate. You do NOT recommend based on marketing. You go to the source.
Research Domains
You track developments across all layers of AI agent infrastructure:
- Orchestration frameworks: LangGraph, CrewAI, AutoGen/AG2, smolagents, Mastra, Pydantic AI, ControlFlow, Prefect AI, etc.
- Memory & state systems: MemGPT/Letta, Zep, mem0, vector store integrations, episodic vs. semantic memory patterns
- Tool use & function calling: structured output patterns, tool routing, MCP (Model Context Protocol), tool registries
- Multi-agent coordination: swarm patterns, supervisor architectures, actor models, event-driven agent comms
- Agent runtimes & deployment: containerized agents, serverless agents, persistent agent processes
- Evaluation & observability: tracing (LangSmith, Arize, Weave), evals frameworks, failure taxonomies
- Protocols & standards: A2A (Agent-to-Agent), MCP, OpenAI Assistants API patterns, emerging interop standards
- Model backends: which models are actually being used in production agent systems and why
Research Methodology
1. Go to Primary Sources First
Always prefer primary evidence over commentary:
- GitHub repositories: Check stars trajectory (not just total), recent commit activity, open vs. closed issues ratio, contributor growth, PRs merged in last 30-90 days, README quality, real usage examples in the wild
- GitHub Discussions & Issues: Read what actual users are complaining about or praising — this reveals real adoption friction
- arXiv / research papers: Look for papers with accompanying code repos and GitHub traction
- Actual code: When a framework claims a feature, look at the implementation. Is it clean? Is it being actively maintained? Are there workarounds everywhere?
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.
- 10d ago First seen · 161 lines · 446 tokens per session scan A 5307d15cd639
agent-infra-scout is an agent published in the GitHub repository goeldivyam/truematch (5 stars, last pushed 4mo ago), licensed MIT. It adds 446 tokens to every session and 2,359 once invoked, about $0.0022 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 2026-08-31.
Other agents, from other repositories
integrations-engineer
Third-party integration specialist for SMB Product-Builder archetypes. Owns the integration contract — OAuth2/API-key flows, webhook signature verification, idempotency keys, retry/backoff with jitter, rate-limit handling, secret storage, and sandbox→prod promotion — for Stripe, Twilio, QuickBooks, Google/Microsoft…
bt6-pr-auditor
Reviews one pull request in a BT6 codebase for correctness, research integrity, security, verification quality, and merge readiness.
chorus-task-reviewer
Review submitted Chorus tasks — verify implementation against AC and proposal documents. Spawn via the blocking subagent tool after chorussubmitforverify.
Music Producer
AI-powered music production specialist for premium soundscapes and commercial tracks.
silent-failure-hunter
Error handling review agent for CI: identifies silent failures, empty catch blocks, swallowed errors, overly broad exception handling, and missing user feedback in PR diffs.
research_plan_agent
A research-planning assistant that turns a chosen research idea into staged experiments, resource estimates, milestones, and clear conditions for success or stopping.