Eval Inline Fact — Data & AI agent for Claude Code
Eval arm inline-fact (Q4 — the CALCULATED vs MEASURED line reduced to its FACT half: what the quantities are, no guidance on what to output).
How to install Eval Inline Fact
Installs to ~/.claude/agents/davegold-mcp-metadata-demo-eval-inline-fact.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/DaveGold/mcp-metadata-demo/HEAD/.claude/agents/eval-inline-fact.md -o ~/.claude/agents/davegold-mcp-metadata-demo-eval-inline-fact.md Restart Claude Code, or start a new session, for it to be picked up.
What Eval Inline Fact does
name: eval-inline-fact description: Eval arm inline-fact (Q4 — the CALCULATED vs MEASURED line reduced to its FACT half: what the quantities are, no guidance on what to output). Answers one building-data question using ONLY the eval-inline-fact MCP server. Not for general use — spawned by the run-eval skill. tools: mcp__eval-inline-fact__get_building_profile, mcp__eval-inline-fact__get_weather_context, mcp__eval-inline-fact__render_chart, mcp__eval-inline-fact__render_table, mcp__eval-inline
Alternatives in Data & AI
- Omd Slop Auditor — 실제 제품 route와 코드를 함께 읽어 context-free UI/copy pattern cluster를 rule ID·line ref·route evidence로 진단한다 465 ★
- Reverse SQL Function Analyst — Spécialiste des FONCTIONS SQL (scalaires, inline, table) du reverse base de données 193 ★
- Blind Fact Check — Zero-context blind review of a single chapter for factual errors and AI-template traces 90 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Eval Inline Instruction
Eval arm inline-instruction (Q4 — the CALCULATED vs MEASURED line reduced to its INSTRUCTION half: what to out
Eval Inline Conditional
Eval arm inline-conditional (Q2 — INTERPRETATION pruned to the record: matching berekeningstype branch plus no
Animation Writer
Writer for a narrated AI animated short film series. Picks an episode topic against the object, half-knowledge
Opus Answer
A simple question-answering agent that responds naturally like a standard LLM. No special research context, no
Question Forger
Generates evaluation questions with gold answers and gold evidence from real graph paths and from the syntheti
Danger Zones - Always Flag These:
pool size reduced (can cause connection starvation) - pool size dramatically increased (can overload database)
Related Skills
Signal Output Style
Signal: a Claude Code output style that answers first, sizes replies to the question, and makes every number c
40+ Claude Code Tips
Tips for getting the most out of Claude Code, including a custom status line script, cutting the system prompt
Craft Notebook
Low-ceremony capture for ideas (half-formed, want to mature into stories), todos (concrete actions), and notes