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.
npx skills add VoDaiLocz/kilo-kit-mcp --skill grounded-research-benchmarkgit clone --depth 1 https://github.com/VoDaiLocz/kilo-kit-mcpWrote 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/skills/vodailocz/kilo-kit-mcp/grounded-research-benchmark)<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.
<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>- NVIDIA SkillSpector pass
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 | $0.00071 | $0.01135 |
| Opus 5 | $0.00036 | $0.00567 |
| Sonnet 5 | $0.00014 | $0.00227 |
| Haiku 4.5 | $0.00007 | $0.00113 |
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.
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:
- Thiết kế hoặc thay đổi kiến trúc quan trọng (Database schema, State Machine, Caching, IPC, Subagents).
- Tác vụ có độ bất định cao (Confidence Score < 0.70).
- Đã thử sửa lỗi 2 lần liên tiếp (
failureStreak >= 2) nhưng test vẫn thất bại. - 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)
- Gọi
kilo_memory_reporthoặc đọc các gợi ý tại Gate 1 để lấy danh sách:- Các
memory_factsđang có hiệu lực. - Các bài học kinh nghiệm (
learning_reflections) của các session trước.
- Các
- Trích xuất các cạm bẫy đã được ghi nhận (
wrongPathsEncountered) để không lặp lại.
Phase 2: External Grounding & Low-Confidence Escalation
- Đánh giá sơ bộ độ tin cậy (Confidence Evaluation):
- Nếu chưa nắm rõ thư viện / API, BẮT BUỘC kích hoạt Research Subagent (
invoke_subagentvới roleresearch). - 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.
- Nếu chưa nắm rõ thư viện / API, BẮT BUỘC kích hoạt Research Subagent (
Phase 3: Cognitive Triangulation & Atomic Persistence (Gate 2)
- Gọi
kilo_triangulate_researchvới: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.
- Gọi
kilo_grill_planđể phản biện rủi ro Adversarial Red-Team (Inversion, Simplification, Blast Radius). - Gọi
kilo_benchmark_solutionđể so khớp với chuẩn công nghiệp.
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.
- 9d ago First seen · 65 lines · 71 tokens per session scan A 7969e6919fd5
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.
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.
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.
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.
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…
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…
memb-skill
BDB local-first long-term memory engine (memB). Use when querying, remembering, or adapting preferences, code architectures, and developer patterns across tasks.