ai-researcher

ai-researcher is an agent for Claude Code from hypnguyen1209/offensive-claude. It costs 28 tokens per session (839 once invoked), scanned A, original, MIT.

An AI and machine-learning research agent that analyzes model designs, training methods, fine-tuning, inference speed, interpretability, and safety.

In plain words
What is it for?
Use it to study transformer and Mamba-style architectures, distributed training, LoRA or RLHF fine-tuning, quantization, KV-cache optimization, mechanistic interpretability, and safety methods.
Why use it?
It gives developers a focused way to investigate how AI models work and how choices affect training, deployment, understanding, and alignment.

Agent for Claude Code

Written for Claude Code: SessionStart hook event. Also seen: model in frontmatter; mentions subagents.

Part of the offensive-claude plugin — 30 skills, 18 commands, 8 agents, 1 hook shipped together

Good fit Use it to study transformer and Mamba-style architectures, distributed training, LoRA or RLHF fine-tuning, quantization, KV-cache optimization, mechanistic interpretability, and safety methods.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/hypnguyen1209/offensive-claude/ai-researcher
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.

Clone the repo
git clone --depth 1 https://github.com/hypnguyen1209/offensive-claude

Made for: Claude Code.

Or install offensive-claude, the plugin that ships this one along with the rest of its 30 skills, 18 commands, 8 agents, 1 hook.

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 ai-researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/hypnguyen1209/offensive-claude/ai-researcher/github.svg)](https://agentmods.dev/agents/hypnguyen1209/offensive-claude/ai-researcher)
Your own site
<a href="https://agentmods.dev/agents/hypnguyen1209/offensive-claude/ai-researcher"><img src="https://agentmods.dev/badge/agents/hypnguyen1209/offensive-claude/ai-researcher/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 ai-researcher

Your own site · 80×15
<a href="https://agentmods.dev/agents/hypnguyen1209/offensive-claude/ai-researcher"><img src="https://agentmods.dev/badge/agents/hypnguyen1209/offensive-claude/ai-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 28 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 839 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 $0.00028 $0.00839
Opus 5 $0.00014 $0.00419
Sonnet 5 $0.00006 $0.00168
Haiku 4.5 $0.00003 $0.00084

Measured 12d ago against content hash cb2c501e914b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

ai-researcher 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 12d 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.

agents/ai-researcher.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an AI/ML research specialist with deep knowledge of model architectures, training methodologies, and the latest research.

Capabilities

  1. Architecture Analysis — Transformer variants, SSMs (Mamba), MoE, hybrid architectures
  2. Training Optimization — distributed training, FSDP, DeepSpeed, Megatron, mixed precision
  3. Fine-tuning — LoRA, QLoRA, DoRA, full fine-tuning, RLHF, DPO, GRPO
  4. Inference Optimization — quantization (GPTQ, AWQ, GGUF), speculative decoding, KV cache optimization
  5. Interpretability — mechanistic interp, sparse autoencoders, activation patching, causal tracing
  6. Safety & Alignment — constitutional AI, guardrails, red-teaming, RLHF/DPO alignment

Research Domains

Model Architecture

  • Attention mechanisms: MHA, GQA, MQA, sliding window, linear attention
  • Position encoding: RoPE, ALiBi, YaRN for context extension
  • Normalization: RMSNorm, LayerNorm placement (pre/post)
  • Activation: SwiGLU, GeGLU
  • Mixture of Experts: routing strategies, load balancing, expert parallelism

Training Infrastructure

  • Parallelism: TP, PP, DP, FSDP2, expert parallelism, context parallelism
  • Optimization: AdamW, LION, Sophia, learning rate schedules
  • Scaling laws: Chinchilla, compute-optimal training
  • Data: curriculum learning, data mixing, deduplication, quality filtering

Post-Training

  • RLHF: reward model training, PPO, rejection sampling
  • DPO/SimPO: reference-free preference optimization
  • GRPO: group relative policy optimization
  • Constitutional AI: self-improvement via principles
  • Distillation: teacher-student, progressive distillation

Inference & Deployment

  • Quantization: INT8, INT4, FP8, mixed precision
  • Serving: vLLM (PagedAttention), TensorRT-LLM, SGLang (RadixAttention)
  • Optimization: Flash Attention, continuous batching, speculative decoding
  • Edge deployment: GGUF, CoreML, TFLite

Output Format

For research questions:

  • Current State: What's known and established
  • Key Papers: Relevant citations with findings
  • Implementation: Practical code/config recommendations
  • Trade-offs: Performance vs cost vs quality analysis
  • Open Questions: What remains unsolved

Read the full file on GitHub · 71 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. 12d ago First seen · 71 lines · 28 tokens per session scan A cb2c501e914b

Subscribe to this mod's changes

ai-researcher is an agent published in the GitHub repository hypnguyen1209/offensive-claude (357 stars, last pushed 25d ago), licensed MIT. It adds 28 tokens to every session and 839 once invoked, about $0.0001 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 2026-08-30.

Related

Other agents, from other repositories

ai-webapi-integration

Use this agent when the user needs to integrate one of the Power Pages generative-AI summarization APIs into their frontend code. The agent supports two APIs: 1. Search Summary — POST /api/search/v1.0/summary 2. Data Summarization — POST /api/summarization/data/v1.0/ ( )?$select=...&$expand=... Data Summarization can…

microsoft/power-platform-skills · 309 tokens

ai-webapi-settings-architect

Use this agent when the user wants to configure Power Pages generative-AI summarization site settings (search summary, data summarization, case-page Copilot preset), enable the Summarization APIs, or register maker-defined prompts for the /api/summarization/data/v1.0/ endpoint. Trigger examples: "enable data…

microsoft/power-platform-skills · 240 tokens

ai-eval-engineer

Builds and maintains the eval pipeline for ai-system / agent-product archetypes. Outputs tests/eval/EVAL-.md files (golden citation, refuse-when-uncertain, output schema, prompt injection, cost-overrun, cross-user isolation). Runs regression on every prompt or model change. Detects drift.

avelikiy/great_cto · 71 tokens

data-platform-reviewer

Data-platform pre-implementation reviewer. Outputs threat model TM-{slug}.md and signs off retention + lineage decisions before senior-dev claims tasks.

avelikiy/great_cto · 33 tokens

geo-routing-engineer

Geospatial and routing specialist for Product-Builder products with maps, scheduling-by-location, or vehicle routing (route-optimization in logistics, dispatch in home services, field-booking). Owns the routing contract — geocoding, the VRP/routing model (constraints, objective), maps/distance-matrix provider…

avelikiy/great_cto · 112 tokens

mlops-reviewer

MLOps / model lifecycle pre-implementation reviewer. Outputs threat model TM-{slug}.md and signs off training-pipeline + serving-strategy decisions before senior-dev claims tasks.

avelikiy/great_cto · 41 tokens