The package manager for agent skills and context
Versioned, evaluated skills and context for agentic software development.
npx tessl search
Used by agentic developers at
Context can drive up to 3.3X improvement in agents use of over 300 libraries
How good is your skill? Put it to the test
Evaluate any skill against structured best practices for descriptions and content.
EVALUATE YOUR SKILLMake agents successful in your environment
Agents don’t know how to develop in your organization, they need to be onboarded. Turn your APIs, libraries, and conventions into agent-usable skills, docs and rules, so agents stop guessing and start behaving like experienced team members.
- Version-matched OSS and internal APIs
- Correct imports, calls, and constraints
- Fewer retries and review cycles
Evaluate what works in real-world scenarios
Not all context is equal, and mistakes can mislead agents or overwhelm context windows. Evaluate and optimize your skills by running agents through real world scenarios, and testing changes to avoid regressions over time.
- Repeatable task evaluations
- Regression detection as skills, agents and models evolve
- Learn if your context helps or hurts agent performance
Create skills once. Use them across all agents and models
Tessl gives you a single source of truth for skills and context, reusable across agents, models, and development environments without duplication or drift.
- Avoid lock-in with universally compatible context
- Consistent behavior across agents
- Collaborate on context with your team and agents
Keep your agent on the rails with better context. Discover thousands of evaluated skills in the Tessl Registry.
Don’t take our word for it
Latest Articles
Explore our guides and resources to understand key concepts relating to Al agents.
Our AI is the bright kid with no manners, part 2
The article explores how an AI's performance drastically drops when deprived of context, highlighting the importance of behavioral training for effective OSS contributions.
Read more
Our AI is the bright kid with no manners, part 1
An AI agent excelled in writing code but struggled with open-source etiquette. After implementing social skills, its acceptance rate rose from 15% to 99%.
Read more
Anthropic tests ‘auto dream’ to clean up Claude Code's memory
Anthropic is testing 'auto dream' for Claude Code to manage memory by reviewing and rewriting stored context, addressing issues of stale or conflicting information.
Read more
Stop guessing whether your Skill works: skill-optimizer measures and improves it
Skill-optimizer evaluates and enhances AI skills by running them through a judge-scored eval pipeline, providing measurable improvements and insights into skill performance.
Read more
Claude Code gets ‘auto mode’ to cut approval fatigue
Claude Code introduces 'auto mode' to balance between constant approvals and full bypass, using a classifier to screen actions for potential risks.
Read more
How Anthropic is turning Claude into an ‘always-on’ agent — and what it learned from OpenClaw
Anthropic is turning up the dial on its efforts to make Claude a system that can carry out tasks without constant input, introducing a trio of new features this past week that let the model run in the background, act across a user’s machine, and be controlled remotely.
Read more
Get started in seconds
curl -fsSL https://get.tessl.io | shOr explore skills and context in the Registry.
