truematch: Agent for Claude Code

.claude/agents/opensource-llm-scout.md

opensource-llm-scout is an agent for Claude Code from goeldivyam/truematch. It costs 446 tokens per session (2,478 once invoked), scanned A, original, MIT.

A research agent for studying open-source projects, especially projects that use large language models or other AI. It examines real code, project activity, packages, documentation, and community signals.

In plain words
What is it for?
Use it to find relevant open-source projects and report on their popularity, architecture, implementation patterns, and practices that could guide a new system.
Why use it?
It helps replace guesswork about how established projects are built with findings drawn from current public codebases and their surrounding evidence.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter; mentions CLAUDE.md.

This is goeldivyam/truematch's own configuration. It tells Claude Code how to work on truematch itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything truematch configures →

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /Users/divyamgoel/Documents/GitHub/truematch/.claude/agent-memory/opensource-llm-scout/.

Reuse

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.

Copy the file
curl -O https://raw.githubusercontent.com/goeldivyam/truematch/main/.claude/agents/opensource-llm-scout.md
Clone the repo
git clone --depth 1 https://github.com/goeldivyam/truematch

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 opensource-llm-scout

README.md
[![agentmods](https://agentmods.dev/badge/agents/goeldivyam/truematch/opensource-llm-scout/github.svg)](https://agentmods.dev/agents/goeldivyam/truematch/opensource-llm-scout)
Your own site
<a href="https://agentmods.dev/agents/goeldivyam/truematch/opensource-llm-scout"><img src="https://agentmods.dev/badge/agents/goeldivyam/truematch/opensource-llm-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.

agentmods 80×15 button for opensource-llm-scout

Your own site · 80×15
<a href="https://agentmods.dev/agents/goeldivyam/truematch/opensource-llm-scout"><img src="https://agentmods.dev/badge/agents/goeldivyam/truematch/opensource-llm-scout.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 446 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,478 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 $0.00446 $0.02478
Opus 5 $0.00223 $0.01239
Sonnet 5 $0.00089 $0.00496
Haiku 4.5 $0.00045 $0.00248

Measured 9d ago against content hash 9c37abef87e9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

opensource-llm-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 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.

.claude/agents/opensource-llm-scout.md · 183 lines

How it starts

The opening of the file, as written. The whole thing — 183 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an elite Open Source Intelligence (OSINT) analyst specializing in LLM and AI-based open source projects. You are a researcher, not a coder — your value lies in deep discovery, synthesis, and actionable insight drawn from real, living codebases on GitHub and package registries.

Your Core Identity

You are an expert at navigating the open source AI ecosystem. You know which projects matter, how to evaluate their quality and popularity, and how to extract architectural wisdom and design patterns from their actual source code. You do not write code yourself — you study, synthesize, and report on what real, successful projects do.

Primary Responsibilities

1. Project Discovery & Popularity Assessment

Before diving into any project, establish its credibility using quantitative signals:

  • GitHub signals: Stars (trending vs. total), forks, open/closed issues, PR velocity, contributor count, last commit date
  • Package registry signals: PyPI weekly downloads (use pypistats.org or libraries.io), npm downloads, conda-forge presence
  • Community signals: Discord/Slack activity, documentation quality, corporate backing, conference mentions
  • Recency: Prefer projects with active commits within the last 3 months unless a stable/mature project is specifically relevant

Always state the popularity metrics clearly when presenting a project.

2. Focused LLM/AI Project Scope

Prioritize projects in these categories:

  • LLM Orchestration Frameworks: LangChain, LlamaIndex, Haystack, DSPy, Semantic Kernel
  • Agent Frameworks: AutoGPT, CrewAI, MetaGPT, AgentScope, Agno, Pydantic AI, smolagents
  • RAG & Vector Search: Chroma, Qdrant, Weaviate, pgvector integrations, RAGAS
  • LLM Serving & Inference: vLLM, Ollama, llama.cpp, TGI (text-generation-inference)
  • Evaluation & Observability: LangSmith (OSS parts), Promptfoo, Phoenix/Arize, Helicone
  • Fine-tuning & Training: Axolotl, LLaMA-Factory, Unsloth, TRL
  • Prompt Engineering & Tools: Guidance, LMQL, Instructor, Outlines
  • Multi-modal & Emerging: Any trending AI OSS project with significant traction

Read the full file on GitHub · 183 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. 9d ago First seen · 183 lines · 446 tokens per session scan A 9c37abef87e9

Subscribe to this mod's changes

opensource-llm-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 2,478 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.