grounded-research-benchmark

grounded-research-benchmark is a skill for Claude Code, Codex from VoDaiLocz/kilo-kit-mcp. It costs 71 tokens per session (1,135 once invoked), scanned A, original, Apache-2.0.

A research process for important architecture decisions that combines stored project knowledge, common open-source patterns, and project constraints.

In plain words
What is it for?
Use it to investigate databases, state machines, caching, communication between processes, or subagent designs before choosing an architecture.
Why use it?
It reduces guesswork when a design is uncertain, a new library is unfamiliar, or repeated fixes have not solved a problem.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions subagents.

Good fit Use it to investigate databases, state machines, caching, communication between processes, or subagent designs before choosing an architecture.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/vodailocz/kilo-kit-mcp/grounded-research-benchmark
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.

Any agent
npx skills add VoDaiLocz/kilo-kit-mcp --skill grounded-research-benchmark
Clone the repo
git clone --depth 1 https://github.com/VoDaiLocz/kilo-kit-mcp

Made for: Claude Code, Codex.

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 grounded-research-benchmark

README.md
[![agentmods](https://agentmods.dev/badge/skills/vodailocz/kilo-kit-mcp/grounded-research-benchmark/github.svg)](https://agentmods.dev/skills/vodailocz/kilo-kit-mcp/grounded-research-benchmark)
Your own site
<a href="https://agentmods.dev/skills/vodailocz/kilo-kit-mcp/grounded-research-benchmark"><img src="https://agentmods.dev/badge/skills/vodailocz/kilo-kit-mcp/grounded-research-benchmark/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 grounded-research-benchmark

Your own site · 80×15
<a href="https://agentmods.dev/skills/vodailocz/kilo-kit-mcp/grounded-research-benchmark"><img src="https://agentmods.dev/badge/skills/vodailocz/kilo-kit-mcp/grounded-research-benchmark.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 71 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,135 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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 $0.00071 $0.01135
Opus 5 $0.00036 $0.00567
Sonnet 5 $0.00014 $0.00227
Haiku 4.5 $0.00007 $0.00113

Measured 9d ago against content hash 7969e6919fd5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

grounded-research-benchmark 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.

skills/grounded-research-benchmark/SKILL.md · 65 lines

How it starts

The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Grounded Research & Cognitive Benchmark Skill

1. Overview & Philosophy

Khi đối mặt với một bài toán kiến trúc mới, công nghệ lạ, hoặc khi độ tin cậy thấp, Agent KHÔNG ĐƯỢC PHÉP ĐOÁN MÒ (Zero Guesswork / Zero Hallucination). Skill này quy chuẩn hóa quy trình Cognitive Triangulation: $$\text{Kiến Trúc Tối Ưu} = \mathcal{F}(\text{Kinh Nghiệm Nội Bộ SQLite},\ \text{Mẫu Chuẩn GitHub 10k+},\ \text{Ràng Buộc Thực Tế Của Dự Án})$$


2. Trigger Conditions

Kích hoạt skill này khi:

  1. Thiết kế hoặc thay đổi kiến trúc quan trọng (Database schema, State Machine, Caching, IPC, Subagents).
  2. Tác vụ có độ bất định cao (Confidence Score < 0.70).
  3. Đã thử sửa lỗi 2 lần liên tiếp (failureStreak >= 2) nhưng test vẫn thất bại.
  4. Tích hợp thư viện mới chưa có trong codebase hiện tại.

3. Mandatory 5-Phase Execution Protocol

Phase 1: Internal Memory Recall (Đọc SQLite không tràn context)

  1. Gọi kilo_memory_report hoặc đọc các gợi ý tại Gate 1 để lấy danh sách:
    • Các memory_facts đang có hiệu lực.
    • Các bài học kinh nghiệm (learning_reflections) của các session trước.
  2. Trích xuất các cạm bẫy đã được ghi nhận (wrongPathsEncountered) để không lặp lại.

Phase 2: External Grounding & Low-Confidence Escalation

  1. Đánh giá sơ bộ độ tin cậy (Confidence Evaluation):
    • Nếu chưa nắm rõ thư viện / API, BẮT BUỘC kích hoạt Research Subagent (invoke_subagent với role research).
    • Subagent tra cứu GitHub (Repo 10k+ stars), tài liệu llms.txt, và AST Signatures trong sandbox riêng.
    • Trả về bản tóm tắt tinh hoa (< 500 tokens) cho Agent chính, tránh làm nổ context window.

Phase 3: Cognitive Triangulation & Atomic Persistence (Gate 2)

  1. Gọi kilo_triangulate_research với:
    • internalMemoryLearned: Bài học cũ từ SQLite.
    • externalGroundingPatterns: Mẫu thiết kế chuẩn từ GitHub.
    • dagOptions: So sánh cụ thể ít nhất 2-3 phương án:
      • Option A: Cách cũ trong DB (hoặc giữ nguyên hiện trạng).
      • Option B: Porting nguyên mẫu từ GitHub.
      • Option C: Tích hợp tinh hoa (Triangulated Synthesis) tối ưu cho dự án.
    • chosenOption: Phương án được chọn kèm lý do.
    • confidenceScore: Điểm tin cậy [0.0 - 1.0]. Nếu < 0.70, tool tự động kích hoạt cờ cảnh báo Research Escalation.
  2. Gọi kilo_grill_plan để phản biện rủi ro Adversarial Red-Team (Inversion, Simplification, Blast Radius).
  3. Gọi kilo_benchmark_solution để so khớp với chuẩn công nghiệp.

Read the full file on GitHub · 65 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 · 65 lines · 71 tokens per session scan A 7969e6919fd5

Subscribe to this mod's changes

grounded-research-benchmark is a skill published in the GitHub repository VoDaiLocz/kilo-kit-mcp (26 stars, last pushed 4d ago), licensed Apache-2.0. It adds 71 tokens to every session and 1,135 once invoked, about $0.0004 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 2026-09-03.

Related

Other skills, from other repositories

llm-structured-output

Get reliable JSON, enums, and typed objects from LLMs using responseformat, tooluse, and schema-constrained decoding across OpenAI, Anthropic, and Google APIs.

hybridlabor-api/bdb-dev-optimized-agent-skills · 42 tokens

drizzle-orm-expert

Expert in Drizzle ORM for TypeScript — schema design, relational queries, migrations, and serverless database integration. Use when building type-safe database layers with Drizzle.

hybridlabor-api/bdb-dev-optimized-agent-skills · 41 tokens

landing-page-generator

Generates high-converting Next.js/React landing pages with Tailwind CSS. Uses PAS, AIDA, and BAB frameworks for optimized copy/components (Heroes, Features, Pricing). Focuses on Core Web Vitals/SEO.

hybridlabor-api/bdb-dev-optimized-agent-skills · 50 tokens

tdmcp-pipeline

Orchestrates the tdmcp feature team end-to-end: design/wireframe → build → integrate → QA → release. Use whenever the user wants to build, implement, develop, ship, or add one or more tdmcp features/tools/effects/controls/AI-prompts for TouchDesigner, or to run them through a coordinated design→develop→QA→deploy…

hybridlabor-api/bdb-dev-optimized-agent-skills · 128 tokens

startcycle-graph-user

Use when a task needs a small, throwaway multi-agent fan-out (a couple of parallel workers plus a review pass) in ANY project, without the full /startcycle-graph contract — no .agents/graph.md bootstrap, no state.json schema, no persistent files. Model-tiered by role (opus plan, sonnet review, haiku or an external CLI…

hybridlabor-api/bdb-dev-optimized-agent-skills · 102 tokens

memb-skill

BDB local-first long-term memory engine (memB). Use when querying, remembering, or adapting preferences, code architectures, and developer patterns across tasks.

hybridlabor-api/bdb-dev-optimized-agent-skills · 36 tokens