{"data":{"slug":"emcie-co-parlant","name":"parlant","tagline":"Build reliable customer-facing AI agents with Parlant: an interaction control harness optimized for controlled, consistent, and predictable LLM interactions.","github_url":"https://github.com/emcie-co/parlant","owner":"emcie-co","repo":"parlant","owner_avatar_url":"https://avatars.githubusercontent.com/u/160175171?v=4","primary_language":"Python","stars":18253,"forks":1551,"topics":["ai-agents","ai-alignment","customer-service","customer-success","gemini","genai","hacktoberfest","llama3","llm","openai","python"],"archived":false,"github_pushed_at":"2026-07-12T19:40:55+00:00","maintenance_label":"Steady","stars_delta_30d":74,"url":"https://www.graphcanon.com/tools/emcie-co-parlant","markdown_url":"https://www.graphcanon.com/tools/emcie-co-parlant.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/emcie-co-parlant","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=emcie-co-parlant","description":"Build reliable customer-facing AI agents with Parlant: an interaction control harness optimized for controlled, consistent, and predictable LLM interactions.","homepage_url":"https://www.parlant.io","license":"Apache-2.0","open_issues":42,"watchers":103,"ai_summary":"Parlant is a tool designed to create and manage AI agents for customer service purposes. It allows developers to define conditions under which specific agent actions are triggered or excluded, ensuring that responses are appropriate and in line with predefined guidelines.","readme_excerpt":"## Getting started\n\n```bash\npip install parlant\n```\n\n```python\nimport parlant.sdk as p\n\nasync with p.Server():\n    agent = await server.create_agent(\n        name=\"Customer Support\",\n        description=\"Handles customer inquiries for an airline\",\n    )\n\n    # Evaluate and call tools only under the right conditions\n    expert_customer = await agent.create_observation(\n        condition=\"customer uses financial terminology like DTI or amortization\",\n        tools=[research_deep_answer],\n    )\n\n    # When the expert observation holds, always respond\n    # with depth. Set the guideline to automatically match\n    # whenever the observation it depends on holds...\n    expert_answers = await agent.create_guideline(\n        matcher=p.MATCH_ALWAYS,\n        action=\"respond with technical depth\",\n        dependencies=[expert_customer],\n    )\n\n    beginner_answers = await agent.create_guideline(\n        condition=\"customer seems new to the topic\",\n        action=\"simplify and use concrete examples\",\n    )\n\n    # When both match, beginners wins. Neither expert-level\n    # tool-data nor instructions can enter the agent's context.\n    await beginner_answers.exclude(expert_customer)\n```\n\nFollow the **[5-minute quickstart](https://www.parlant.io/docs/quickstart/installation)** for a full walkthrough.\n\n---\n\n## License\n\nApache 2.0 — free for commercial use.\n\n---\n\n<div align=\"center\">\n\n**[Try it now](https://www.parlant.io/docs/quickstart/installation)** &bull; **[Join Discord](https://discord.gg/duxWqxKk6J)** &bull; **[Read the docs](https://www.parlant.io/)**\n\nBuilt by the team at **[Emcie](https://emcie.co)**\n\n</div>","github_created_at":"2024-02-15T20:16:15+00:00","created_at":"2026-07-07T17:40:46.469892+00:00","updated_at":"2026-08-20T00:01:10.29706+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"}],"tags":[{"slug":"ai-agents","name":"ai-agents"},{"slug":"ai-alignment","name":"ai-alignment"},{"slug":"customer-service","name":"customer-service"},{"slug":"customer-success","name":"customer-success"},{"slug":"gemini","name":"gemini"},{"slug":"genai","name":"genai"},{"slug":"hacktoberfest","name":"hacktoberfest"},{"slug":"llama3","name":"llama3"}],"trust":{"provenance":{"is_fork":false,"github_id":758197374,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-08-20T00:01:09.529Z","maintenance":{"label":"Steady","score":60,"methodology":"github_public_v1","releases_90d":0,"days_since_push":38,"last_release_at":"2026-04-28T06:32:34Z","stars_delta_30d":74,"open_issues_delta_30d":3},"security_summary":{"status":"findings","scanner":"mcp_manifest@v1","low_count":2,"high_count":0,"last_scan_at":"2026-07-11T11:17:34.370Z","medium_count":0,"scan_profile":"mcp_manifest","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-08-20T00:01:09.981Z"},"has_cli":{"value":true,"source":"pyproject.toml:[project.scripts]","observed_at":"2026-08-20T00:01:09.981Z"},"languages":{"value":["python"],"source":"github.language+pyproject.toml","observed_at":"2026-08-20T00:01:09.981Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-08-20T00:01:09.981Z"}},"decision_facts":{"hosting":null,"pricing":{"model":"freemium","summary":"The open-source version offers community-supported functionalities which are free to use but might lack advanced support services."},"requirements":{"min_ram_gb":null,"requires_docker":false},"constraints":{"min_ram_gb":null,"pricing_model":"freemium","requires_docker":false},"when_to_use":["When you need precise control over the actions of AI agents based on specific conditions or observations.","For scenarios where responses must adapt to user behavior, such as distinguishing between technical experts and beginners among customers.","If your application requires that certain guidelines (like simplifying explanations) are only applied when a condition is met, ensuring consistency in how customer queries are handled."],"when_not_to_use":["When the requirements for agent interactions do not demand granular rules based on observable user behavior or conditions, making this level of control unnecessary.","In cases where the flexibility and spontaneous responses from general-purpose LLMs are preferred without being restricted by controlled guidelines.","If your project involves minimal interaction types that can be managed with less sophisticated frameworks, thus negating the need for Parlant's advanced condition setting capabilities."],"source":"enrich:decision_facts","observed_at":"2026-07-11T14:30:04.668Z"},"constraint_facets":{"min_ram_gb":null,"pricing_model":"freemium","requires_docker":false},"decision_summary":[{"label":"Pricing","value":"freemium - The open-source version offers community-supported functionalities which are free to use but might lack advanced support services."},{"label":"Adopt for","value":"Parlant is a specialized interaction control harness designed to ensure that AI agents, particularly those in customer service roles, behave predictably and align with predefined rules."},{"label":"License detail","value":"Parlant is available under the Apache-2.0 license, allowing for unrestricted commercial use as long as proper attribution is given."}]}}