agent-trajectory is a Python library that standardizes transcripts from various native AI agent frameworks into a common trajectory format. It enables consistent data processing, training, and analysis across different agent implementations. Primarily used by AI researchers and developers building or evaluating agentic systems.
agent-trajectory sits in PulseGate's AI & ML category. Inconsistent formats when working with transcripts from different AI agent frameworks. agent-trajectory is an open-source project aimed at AI developers. The project is open source (Apache-2.0). agent-trajectory is available on the command line.
Letta AI builds and maintains agent-trajectory, and the product first shipped in 2026. The project is developed in the open on GitHub with 192 stars and 79 commits in the last 90 days.
Latest indexed changes and source events
agent-trajectory verified by the PulseGate indexer
Other apps tracked under the same category.