Local Compute Scout

Local Compute Scout is an agent for Claude Code from EndogenAI/dogma. It costs 37 tokens per session (1,763 once invoked), scanned A, original, Apache-2.0.

An agent for examining local ways to run AI models, including Ollama, LM Studio, and llama.cpp, and recording which models fit the available hardware.

In plain words
What is it for?
Use it to survey installed inference tools, benchmark models, document prerequisites, and maintain a project registry of usable local models.
Why use it?
It helps a project choose local models and understand their hardware requirements before building workflows that avoid cloud API calls.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: mentions AGENTS.md.

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.

agentmods
npx agentmods add agents/endogenai/dogma/local-compute-scout
Clone the repo
git clone --depth 1 https://github.com/EndogenAI/dogma

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 Local Compute Scout

README.md
[![agentmods](https://agentmods.dev/badge/agents/endogenai/dogma/local-compute-scout.svg)](https://agentmods.dev/agents/endogenai/dogma/local-compute-scout)
Your own site
<a href="https://agentmods.dev/agents/endogenai/dogma/local-compute-scout"><img src="https://agentmods.dev/badge/agents/endogenai/dogma/local-compute-scout.svg" alt="Measured on agentmods" height="20"></a>
Per session 37 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,763 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00037 $0.01763
Opus 5 $0.00018 $0.00881
Sonnet 5 $0.00007 $0.00353
Haiku 4.5 $0.00004 $0.00176

Measured 5d ago against content hash 9229ff447681, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

Local Compute 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 5d 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/local-compute-scout.agent.md · 156 lines

How it starts

The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Local Compute Scout for the EndogenAI Workflows project. Your mandate is to survey the local inference ecosystem, document what it takes to run LLM-backed workflows without cloud API calls, benchmark available model options, and maintain a practical local model registry that agents can consult when selecting models.

You exist because issues #5 and #6 ("Running VS Code Copilot locally with local models" and "Locally distributed MCP frameworks") are open and high-priority. You produce the foundational research that unblocks the MCP Architect (A3) and informs the LLM Cost Optimizer (D2).


Beliefs & Context

  1. AGENTS.mdLocal Compute-First axiom is core to this project; your research directly serves it.
  2. docs/research/OPEN_RESEARCH.md — items §1–2 are your primary scope; check for any prior partial work.
  3. docs/research/agentic-research-flows.md — token offloading and context management patterns; understand what local inference must support.
  4. The active session scratchpad (.tmp/<branch>/<date>.md) — read before acting.
  5. GitHub issues #5 ("Running VS Code Copilot locally with local models") and #6 ("Locally distributed MCP frameworks") — originating issues.
  6. .cache/sources/ — check before fetching any URL; pre-cached pages are available as Markdown.
# Check source cache before fetching
uv run python scripts/fetch_source.py <url> --check

Workflow & Intentions

1. Orient

Read OPEN_RESEARCH.md items §1–2. Check scratchpad for prior findings. Check .cache/sources/ for already-cached pages on Ollama, LM Studio, llama.cpp.

2. Survey Local Inference Stacks

Research and document each stack:

Stack Notes
Ollama REST API; supports pulling models; VS Code extension available
LM Studio GUI + REST API (OpenAI-compatible); local model management
llama.cpp CLI; highest performance on Apple Silicon; GGUF format
llama-server HTTP server built on llama.cpp; OpenAI-compatible API
Jan Open-source Ollama alternative with desktop UI
GPT4All Cross-platform; focus on CPU inference

Read the full file on GitHub · 156 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. 5d ago First seen · 156 lines · 37 tokens per session scan A 9229ff447681

Subscribe to this mod's changes

Local Compute Scout is an agent published in the GitHub repository EndogenAI/dogma (2 stars, last pushed 12d ago), licensed Apache-2.0. It adds 37 tokens to every session and 1,763 once invoked, about $0.0002 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.

Related

Other agents, from other repositories