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 jxoesneon/Ciel --skill research-and-intelligencegit clone --depth 1 https://github.com/jxoesneon/CielWrote 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/jxoesneon/ciel/research-and-intelligence)<a href="https://agentmods.dev/skills/jxoesneon/ciel/research-and-intelligence"><img src="https://agentmods.dev/badge/skills/jxoesneon/ciel/research-and-intelligence/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/jxoesneon/ciel/research-and-intelligence"><img src="https://agentmods.dev/badge/skills/jxoesneon/ciel/research-and-intelligence.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 | $0.00023 | $0.00550 |
| Opus 5 | $0.00012 | $0.00275 |
| Sonnet 5 | $0.00005 | $0.00110 |
| Haiku 4.5 | $0.00002 | $0.00055 |
Grade A, and why
research-and-intelligence 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.
What it actually says
CIEL ADAPTATION: Research & Intelligence (The Knowledge Layer)
This skill formalizes the retrieval and synthesis of external knowledge. it prioritizes cited evidence and automated data collection.
Deep Research Workflow
- Breakdown: Split the core topic into 3-5 sub-questions (Market, Tech, Competitors, Trends).
- Multi-Source Search: Execute searches via Exa and Firecrawl using 3+ keyword variations per question.
- Deep Read: Scrape the top 3-5 key URLs in full. do NOT rely on snippets.
- Synthesize: Produce a report with inline citations and a "Confidence Score" (High/Medium/Low).
Automated Scraping (Agent)
- Collect: Use
requestsorplaywright(for JS-rendered sites). Respectrobots.txt. - Enrich: Batch items (5 per call) to Gemini Flash for scoring and summarization.
- Store: Sync unique items to Notion/Sheets/Supabase. Deduplicate by URL.
- Learn: Use a
feedback.jsonto bias future AI scores based on user "Likes/Dislikes."
Live Documentation (Context7)
- Resolve: Map library names (e.g., "React 19") to specific library IDs early.
- Query: Fetch live code snippets and configuration guides instead of relying on training data.
- Redact: Ensure no API keys or secrets are passed in documentation queries.
Anti-Patterns
- Unsourced Claims: Adding "facts" to a report without a clickable citation.
- Snippet-Only Synthesis: Writing a deep dive based only on 200-character search results.
- Rate-Limit Blindness: Running 50 individual LLM calls for 50 scraped items.
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 · 52 lines · 23 tokens per session scan A ee82b9aee1ae
research-and-intelligence is a skill published in the GitHub repository jxoesneon/Ciel (1 stars, last pushed 5d ago), licensed Apache-2.0. It adds 23 tokens to every session and 550 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 2026-09-03.
Other skills, from other repositories
ralph-prompt-generator
Use when a user has a rough prompt that needs a staged prompt-improvement workflow for $ralph-wiggum-codex, with saved draft/final files and explicit review checkpoints.
llms-txt-generator
Scans your documentation and key pages to create a structured text file optimized for LLM reading.
local-llm-tool
Local LLM execution tool for text generation and chat through Ollama or vLLM endpoints. Use when: running on-prem inference, calling a local GPU model, or summarizing with a self-hosted LLM.
local-llm
A setup and control layer for running language models locally on a Mac with Apple silicon, so data can stay on the computer.
AgentDB Learning Plugins
Create and train AI learning plugins with AgentDB's 9 reinforcement learning algorithms. Includes Decision Transformer, Q-Learning, SARSA, Actor-Critic, and more. Use when building self-learning agents, implementing RL, or optimizing agent behavior through experience.
AgentDB Performance Optimization
Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch operations. Use when optimizing memory usage, improving search speed, or scaling to millions of vectors.