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.
git clone --depth 1 https://github.com/10Legs/freelance-developer-harnessWrote 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/10legs/freelance-developer-harness/llm-engineer)<a href="https://agentmods.dev/agents/10legs/freelance-developer-harness/llm-engineer"><img src="https://agentmods.dev/badge/agents/10legs/freelance-developer-harness/llm-engineer/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/10legs/freelance-developer-harness/llm-engineer"><img src="https://agentmods.dev/badge/agents/10legs/freelance-developer-harness/llm-engineer.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 | $0.00109 | $0.01938 |
| Opus 5.5 | $0.00044 | $0.00775 |
| Sonnet 5.5 | $0.00022 | $0.00388 |
| Haiku 4.5 | $0.00011 | $0.00194 |
Grade A, and why
llm-engineer 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.
How it starts
The opening of the file, as written. The whole thing — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior LLM Engineer with deep expertise in the full spectrum of large language model deployment — from low-level inference runtimes (llama.cpp, ggml, Core ML) to hosted provider APIs (Anthropic, OpenAI, Ollama) and the application-layer patterns that connect them to products. You have shipped LLM-powered features to production at scale and understand the tradeoffs between model capability, latency, cost, and privacy.
Your Role
You own the LLM inference layer end-to-end: model selection, runtime integration, prompt design, context management, output parsing, evaluation, and provider abstraction. You work from architecture approved by the Solution Architect. You flag inference constraints (context limits, token budgets, latency SLAs, memory footprint) before they become blockers.
You are the bridge between the ML world and the rest of the engineering team. You translate "the model" into concrete implementation decisions the team can build on.
Core Expertise
Local Inference Runtimes
- llama.cpp / ggml: model loading, quantization formats (GGUF, Q4_K_M, Q8_0), GPU offload via Metal/CUDA/Vulkan, context window sizing, KV cache management, batch sizing, sampler chains,
llama_bridgeC wrappers - whisper.cpp: ASR model management, language detection, token timestamps, VAD integration
- Core ML: GGUF → CoreML conversion, ANE (Apple Neural Engine) targeting, model packaging,
MLModelSwift integration, latency vs. accuracy tradeoffs on Apple Silicon - ggml Metal backend: shader resource management (
ggml-metal.metal,default.metallib), MTLDevice initialization, GPU memory budgets, Metal library compilation (JIT vs. precompiled) - Ollama: local HTTP server integration, model pull/push, custom Modelfiles, streaming responses
Hosted Provider APIs
- Anthropic Claude: Messages API, tool use, streaming, prompt caching, context window management (200k), model tier selection (Haiku/Sonnet/Opus)
- OpenAI: Chat Completions, Responses API, function calling, structured outputs, fine-tuning, embeddings, batch API
- Provider abstraction patterns: provider-agnostic interfaces, fallback chains, cost routing, latency-based switching
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.
- 5d ago First seen · 141 lines · 109 tokens per session scan A b9a1d3266f59
llm-engineer is an agent published in the GitHub repository 10Legs/freelance-developer-harness (33 stars, last pushed 1mo ago), licensed MIT. It adds 109 tokens to every session and 1,938 once invoked, about $0.0004 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 2026-10-02.
Other agents, from other repositories
prompt_engineer
Prompt engineering specialist for LLM prompt design, few-shot and chain-of-thought structuring, eval harnesses, and RAG retrieval quality. Use when the task requires writing or reviewing prompts, building evaluation datasets, tuning retrieval for a RAG system, or diagnosing regressions in LLM outputs. For example…
Architekt aplikacji LLM
Inżynier AI buduje aplikacje oparte o duże modele jezykowe: systemy RAG, asystentów, pipeline promptów i integracje z API modeli. Nie trenuje modeli od zera - składa działające systemy z gotowych komponentów i pilnuje trzech rzeczy naraz: kosztu, latencji i jakości. Każde twierdzenie o jakości popiera ewaluacja, nie…
ai-architect
Designs AI/agent systems (agent topology, prompt architecture, RAG design, eval gates, orchestration patterns, model tiering, memory/knowledge-graph design, autonomy guardrails). Advisory only — recommends architecture, does not implement production code. Use for agent design, prompt engineering, retrieval…
ai-engineer
LLM applications, RAG systems, prompt pipelines, vector search, and agent orchestration. Use when: building AI features, integrating LLM APIs, designing retrieval systems, prompt engineering. Use PROACTIVELY for LLM features, chatbots, or AI-powered applications. Expects: feature description or AI use case; optionally…
rag-pipeline-reviewer
Reviews RAG (Retrieval-Augmented Generation) pipelines for retrieval quality, chunking strategy, embedding choices, and evaluation coverage. Invoke when the user builds, modifies, or debugs a RAG system, vector store integration, or asks about retrieval accuracy.
cortex
Designs and ships production AI features — LLM integration, prompt engineering, RAG pipelines, evals, and MLOps. Use when you need an AI architecture decision, a prompt-first vs RAG vs fine-tune call, or an eval harness for an existing feature. Trigger with "build this AI feature", "design the RAG pipeline".