Griptape
Modular Python framework for AI agents — 2.6K stars, Apache 2.0, actively maintained.
About
Griptape (2,580 GitHub stars, Apache-2.0) is a modular Python framework for building AI agents: structures (agents, pipelines, workflows), tools, memory, and drivers for any LLM provider.
Actively maintained (pushed Oct 2026). The modular design (separate Drivers for prompts, embeddings, vector stores) suits teams that want provider-swappable components. Smaller community than the majors, but the architecture is clean and the maintenance is steady. For Python teams that want explicit control over agent components.
What it can do
- • Modular agent framework (Structures)\n• Provider-swappable Drivers\n• Tools, memory, and RAG components\n• Pipelines and workflows\n• Multi-agent task management
Setup
- Install:
pip install griptape.\n2. Set your model API key.\n3. Build Structures (Agent/Pipeline/Workflow) with Tools and Memory.\nSee https://docs.griptape.ai for structure guides.
Prerequisites
Python ≥ 3.10; an LLM API key.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (metrics tier). Apache-2.0, actively maintained with 17 releases, and a clean modular architecture. Docked three points: (1) small community (2.6K stars); (2) overlaps heavily with larger frameworks; (3) limited ecosystem. For Python teams wanting explicit component control.
★ 2,580 GitHub stars
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 6.6/10
- Maintenance · recency, releases, issue load
- 8.5/10
- Openness · license + docs
- 9.1/10
- Manual review · sanity read + red-flag check
- 6.0/10
- · 2,580 GitHub stars (log-scaled vs catalog peers)
- · last push 7d ago; 17 releases/12mo; 97 open issues
- · manual review: clean modular architecture; small community; overlaps larger frameworks; license: Apache-2.0
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.