vstorm-co/summarization-pydantic-ai

Context Management processor for Pydantic AI agents, providing LLM-powered summarization or zero-cost sliding window trimming to handle infinite/long-running conversations without context overflow. Supports flexible triggers, safe cutoffs, and custom prompts for efficient AI apps.

56
/ 100
Established

Implements both **Capabilities API integration** (auto-detects model context windows) and **standalone Processor API** for flexible deployment with Pydantic AI. Features real-time token counting via `tiktoken`, intelligent tool output truncation that preserves call/response pairs, and configurable triggers (message count, token count, or context fraction) with multiple keep strategies to balance compression quality against latency and cost.

10 stars and 71,066 monthly downloads. Available on PyPI.

Maintenance 10 / 25
Adoption 15 / 25
Maturity 18 / 25
Community 13 / 25

How are scores calculated?

Stars

10

Forks

2

Language

Python

License

MIT

Last pushed

Feb 27, 2026

Monthly downloads

71,066

Commits (30d)

0

Dependencies

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/agents/vstorm-co/summarization-pydantic-ai"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.