plugin for everyrow

.claude-plugin/plugin.json

@@ -0,0 +1,10 @@
+{
+  "name": "everyrow",
+  "description": "Claude Code plugin for the everyrow SDK - AI-powered data processing utilities for transforming, deduping, merging, ranking, and screening dataframes",
+  "version": "1.0.0",
+  "author": {
+    "name": "FutureSearch"
+  },
+  "repository": "https://github.com/futuresearch/everyrow-sdk",
+  "skills": "./plugin/skills/"
+}

plugin/skills/everyrow-sdk/SKILL.md

@@ -0,0 +1,206 @@
+---
+name: everyrow-sdk
+description: Helps write Python code using the everyrow SDK for AI-powered data processing - transforming, deduping, merging, ranking, and screening dataframes with natural language instructions
+---
+
+# everyrow SDK
+
+The everyrow SDK provides intelligent data processing utilities powered by AI agents. Use this skill when writing Python code that needs to:
+- Deduplicate data using semantic understanding
+- Merge tables using AI-powered matching
+- Rank/score rows based on AI evaluation
+- Screen/filter rows based on research-intensive criteria
+- Run AI agents over dataframe rows
+
+## Installation
+
+```bash
+pip install everyrow
+# or
+uv pip install everyrow
+```
+
+## Configuration
+
+Requires an API key from https://everyrow.io:
+
+```bash
+export EVERYROW_API_KEY=your_api_key_here
+```
+
+## Core Pattern
+
+All operations are async and require a session:
+
+```python
+import asyncio
+from everyrow import create_client, create_session
+
+async def main():
+    async with create_client() as client:
+        async with create_session(client=client, name="My Session") as session:
+            print(f"View session at: {session.get_url()}")
+            # ... use session for operations
+
+asyncio.run(main())
+```
+
+Simplified version (client created automatically):
+
+```python
+from everyrow import create_session
+
+async with create_session(name="My Session") as session:
+    # ... use session
+```
+
+## Operations
+
+### dedupe - Deduplicate data
+
+Remove duplicates using AI-powered semantic matching:
+
+```python
+from everyrow.ops import dedupe
+
+result = await dedupe(
+    session=session,
+    input=dataframe,
+    equivalence_relation="Two entries are duplicates if they represent the same research work, even with different titles or authors listed",
+)
+# result.data contains the deduplicated DataFrame
+```
+
+### merge - Merge tables with AI matching
+
+Join tables where the match criteria requires understanding:
+
+```python
+from everyrow.ops import merge
+
+result = await merge(
+    session=session,
+    task="Match clinical trial sponsors with their parent pharmaceutical companies",
+    left_table=trial_data,
+    right_table=company_data,
+    merge_on_left="sponsor",
+    merge_on_right="company",
+)
+```
+
+### rank - Score and rank rows
+
+Add AI-generated scores to rank rows:
+
+```python
+from everyrow.ops import rank
+
+result = await rank(
+    session=session,
+    task="Score this organization by their contribution to open source AI research (0-100)",
+    input=dataframe,
+    field_name="contribution_score",
+    ascending_order=False,
+)
+```
+
+### screen - Evaluate and filter rows
+
+Evaluate rows based on criteria that requires research:
+
+```python
+from everyrow.ops import screen
+from pydantic import BaseModel
+
+class VendorAssessment(BaseModel):
+    approved: bool
+    risk_level: str  # "low", "medium", "high"
+    security_concerns: str
+    recommendation: str
+
+result = await screen(
+    session=session,
+    task="""Evaluate vendor security and financial stability:
+    1. Have they had data breaches in the past 3 years?
+    2. Are there signs of financial distress?
+    Only approve vendors with low/medium risk.""",
+    input=vendors_df,
+    response_model=VendorAssessment,
+    batch_size=5,
+)
+```
+
+### single_agent - Single input task
+
+Run an AI agent on a single input:
+
+```python
+from everyrow.ops import single_agent
+from pydantic import BaseModel
+
+class Input(BaseModel):
+    company: str
+
+result = await single_agent(
+    session=session,
+    task="Research this company and summarize their main products",
+    input=Input(company="Anthropic"),
+    return_table=False,  # Set True for table output
+)
+```
+
+### agent_map - Batch processing
+
+Run an AI agent across multiple rows:
+
+```python
+from everyrow.ops import agent_map
+from pandas import DataFrame
+
+result = await agent_map(
+    session=session,
+    task="What is the capital of the given country?",
+    input=DataFrame([{"country": "India"}, {"country": "USA"}]),
+    return_table_per_row=False,
+)
+```
+
+## Async Variants
+
+All operations have `_async` variants for background processing:
+
+```python
+from everyrow.ops import rank_async
+
+task = await rank_async(
+    session=session,
+    task="Score organizations",
+    input=dataframe,
+    field_name="score",
+)
+
+# Do other work...
+
+result = await task.await_result(session.client)
+```
+
+## Result Structure
+
+All operations return a result object with:
+- `result.data` - The output DataFrame
+- `result.artifact_id` - ID for the stored artifact
+
+## Effort Levels
+
+Control processing intensity with `effort_level`:
+
+```python
+from everyrow.generated.models import TaskEffort
+
+result = await single_agent(
+    session=session,
+    task="...",
+    input=input_data,
+    effort_level=TaskEffort.LOW,  # or MEDIUM, HIGH
+)
+```