agentjit
PulseGate's liveness check found it on 12 Sep 2026; it is registered on GitHub and PyPI and has been in the index since 12 Sep 2026. How this is checked
agentjit is an Apache-2.0 Python package that compiles multi-step AI agent trajectories into deterministic Python code. It targets developers who want faster execution and lower token usage for repeatable LLM workflows.
Inferred · not functionally tested
Overview
6 featuresPurpose: Reducing the latency and token cost of repeated multi-step LLM workflows.
Inferred · not functionally tested
Audience: AI developers building multi-step agent workflows
Inferred · not functionally tested
Functions: code_generation, agents
Inferred · not functionally tested
Interfaces: API: indicated (inferred, not tested) · MCP: unknown · CLI: indicated (inferred, not tested) · Self-hosting: indicated (inferred, not tested)
Recorded constraints: pricing: open_source · license: Apache-2.0 · platforms: CLI · deployment: cli, self_hosted, api_only
Constraint provenance is unknown; confirm requirements with the publisher.
Record sources: pypi.org · github.com. These links do not verify the individual claims.
In the Frameworks & runtimes space, agentjit takes a focused approach. Inferred · not functionally tested: It focuses on reducing the latency and token cost of repeated multi-step LLM workflows. Inferred · not functionally tested: agentjit is an open-source project aimed at AI developers building multi-step agent workflows. Basis unknown · not verified: agentjit is open source under the Apache-2.0 license. Basis unknown · not verified: It runs on the command line and API, and it can be self-hosted.
eminsk builds and maintains agentjit, and it first shipped in 2026. The project is developed in the open on GitHub with 24 commits in the last 90 days. Inferred · not functionally tested: Among its 6 catalogued features are trajectory compilation, python code generation, and deterministic execution.
Summary written by a language model from the project’s public pages.
Tasks: Inferred · not functionally tested
- Trajectory compilation
- Python code generation
- Deterministic execution
- Workflow optimization
- Token reduction
- Free-threading support
Topics: Inferred · not functionally tested
Built with & integrations
Trust & compliance
Indexing history
4What PulseGate has recorded for this listing
Frequently asked questions about agentjit
- What does agentjit do?
- Inferred · not functionally tested: Agentjit focuses on reducing the latency and token cost of repeated multi-step LLM workflows. It is catalogued under Frameworks & runtimes on PulseGate.
- Who is agentjit for?
- Inferred · not functionally tested: agentjit is an open-source project built for AI developers building multi-step agent workflows.
- Does agentjit have a free plan?
- Basis unknown · not verified: Yes — agentjit is open source under the Apache-2.0 license and free to use.
- What platforms does agentjit run on?
- Basis unknown · not verified: agentjit runs on the command line and API. It can also be self-hosted.
- Is agentjit still active?
- PulseGate's liveness check found it on 12 Sep 2026. Its GitHub repository shows 24 commits in the last 90 days.
- What are alternatives to agentjit?
- Similar projects tracked by PulseGate include ai-agent-task, agentgod, and agent-teams.ai-agent-taskagentgodagent-teams
- Who develops agentjit?
- agentjit is developed by eminsk.
- How long has agentjit been around?
- agentjit first shipped in 2026.
Similar projects
Closest matches by what these projects do