diff --git a/.gitignore b/.gitignore
index 628546d..bb351a0 100644
--- a/.gitignore
+++ b/.gitignore
@@ -48,6 +48,9 @@ venv.bak/
 .DS_Store
 .claude/
 
+# Local documentation
+LOCAL_OPERATIONS.md
+
 # Database files
 *.sqlite
 *.sqlite3
diff --git a/README.md b/README.md
index a3a6b6a..0e483ee 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
 # Mastodon Collector
 
-Collects posts, replies, and mentions from a list of Mastodon accounts and stores them in PostgreSQL. Includes automated toxicity analysis using OpenAI GPT-4o-mini, a web UI for account management, data browsing, and manual review of flagged content, plus JSON/CSV APIs for your analysis pipeline.
+Collects posts, replies, and mentions from a list of Mastodon accounts and stores them in PostgreSQL. Includes automated toxicity analysis using LLM API, a web UI for account management, data browsing, and manual review of flagged content, plus JSON/CSV APIs for your analysis pipeline.
 
 ## Quick Start
 
@@ -38,7 +38,7 @@ Edit `.env` to customize:
 POSTGRES_PASSWORD=collector_secret      # Change for production
 FLASK_SECRET_KEY=change-me-in-production
 POLL_INTERVAL_SECONDS=14400             # Default: 4 hours (14400s)
-OPENAI_API_KEY=sk-...                   # Required for toxicity analysis
+LLM_API_KEY=sk-...                      # Required for toxicity analysis
 ```
 
 ## Toxicity Analysis
@@ -47,7 +47,7 @@ The system includes automated toxicity detection and manual review capabilities:
 
 ### Features
 
-- **Automated Classification**: Uses OpenAI GPT-4o-mini to analyze posts across 12 toxicity dimensions:
+- **Automated Classification**: Uses LLM API to analyze posts across 12 toxicity dimensions:
   - General toxicity, threats, hate speech
   - Racism, antisemitism, islamophobia
   - Sexism, homophobia, ableism
diff --git a/app/analyzer/analyzer.py b/app/analyzer/analyzer.py
index a1c6731..fb52d37 100644
--- a/app/analyzer/analyzer.py
+++ b/app/analyzer/analyzer.py
@@ -1,7 +1,7 @@
 """Main toxicity analysis orchestrator.
 
 Runs as a one-shot batch process: fetches unscored statuses,
-classifies them in batches with GPT-4o-mini, and stores scores in PostgreSQL.
+classifies them in batches with LLM API, and stores scores in PostgreSQL.
 
 Usage:
     python -m app.analyzer
@@ -120,7 +120,7 @@ async def run() -> None:
 
     db = AnalyzerDB(config.database_url)
     classifier = ToxicityClassifier(
-        api_key=config.openai_api_key,
+        api_key=config.llm_api_key,
         model=config.model,
     )
 
diff --git a/app/analyzer/classifier.py b/app/analyzer/classifier.py
index a3de131..07d80ec 100644
--- a/app/analyzer/classifier.py
+++ b/app/analyzer/classifier.py
@@ -1,6 +1,6 @@
-"""OpenAI-powered toxicity classifier.
+"""LLM-powered toxicity classifier.
 
-Sends batches of Mastodon statuses to GPT-4o-mini for multi-category toxicity scoring.
+Sends batches of Mastodon statuses to LLM API for multi-category toxicity scoring.
 Returns a list of dicts of category → score (0.0–1.0).
 """
 
@@ -181,7 +181,7 @@ def parse_batch_response(raw: str, batch_size: int) -> list[ToxicityScores]:
 
 
 class ToxicityClassifier:
-    """Async OpenAI-based toxicity classifier with batch support."""
+    """Async LLM-based toxicity classifier with batch support."""
 
     def __init__(self, api_key: str, model: str = "gpt-4o-mini"):
         self.client = AsyncOpenAI(api_key=api_key)
diff --git a/app/analyzer/config.py b/app/analyzer/config.py
index 64b1cd2..7f0d486 100644
--- a/app/analyzer/config.py
+++ b/app/analyzer/config.py
@@ -11,7 +11,7 @@ class AnalyzerConfig:
     """Configuration for the toxicity analyzer."""
 
     database_url: str
-    openai_api_key: str
+    llm_api_key: str
     model: str = "gpt-4o-mini"
     batch_size: int = 10
     concurrency: int = 5
@@ -28,7 +28,7 @@ class AnalyzerConfig:
         """Load configuration from environment variables."""
         return cls(
             database_url=os.environ["DATABASE_URL"],
-            openai_api_key=os.environ["OPENAI_API_KEY"],
+            llm_api_key=os.environ["LLM_API_KEY"],
             model=os.getenv("ANALYZER_MODEL", "gpt-4o-mini"),
             batch_size=int(os.getenv("ANALYZER_BATCH_SIZE", "10")),
             concurrency=int(os.getenv("ANALYZER_CONCURRENCY", "5")),
diff --git a/docker-compose.yml b/docker-compose.yml
index 06674f8..755fbc6 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -40,7 +40,7 @@ services:
     environment:
       DATABASE_URL: postgresql://collector:${POSTGRES_PASSWORD:-collector_secret}@db:5432/mastodon_collector
       POLL_INTERVAL_SECONDS: ${POLL_INTERVAL_SECONDS:-14400}
-      OPENAI_API_KEY: ${OPENAI_API_KEY}
+      LLM_API_KEY: ${LLM_API_KEY}
     volumes:
       - ./accounts.txt:/app/accounts.txt
     depends_on:
diff --git a/requirements.txt b/requirements.txt
index 735ed42..d05196b 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -5,5 +5,5 @@ sqlalchemy==2.0.36
 requests==2.32.3
 apscheduler==3.10.4
 beautifulsoup4==4.12.3
-openai==1.58.1
+openai==1.58.1  # OpenAI-compatible API client (supports any LLM provider)
 asyncpg==0.30.0