{"data":{"slug":"hidai25-eval-view","name":"eval-view","tagline":"Regression testing for AI agents, snapshots behavior, diffs tool calls, catches regressions in CI","github_url":"https://github.com/hidai25/eval-view","owner":"hidai25","repo":"eval-view","owner_avatar_url":"https://avatars.githubusercontent.com/u/31502796?v=4","primary_language":"Python","stars":134,"forks":24,"topics":["agent-benchmark","agent-evaluation","agentic-ai","ai-agents","anthropic","autogen","cli","crewai","evaluation","langchain-agent","langgraph","llm","mcp","openai-assistants","pytest","python","regression-testing","testing"],"archived":false,"github_pushed_at":"2026-09-05T03:28:43+00:00","maintenance_label":"Active","stars_delta_30d":8,"url":"https://www.graphcanon.com/tools/hidai25-eval-view","markdown_url":"https://www.graphcanon.com/tools/hidai25-eval-view.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/hidai25-eval-view","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=hidai25-eval-view","description":"Regression testing for AI agents. Snapshot behavior,diff tool calls,catch regressions in CI. Works with LangGraph, CrewAI, OpenAI, Anthropic.","homepage_url":"https://evalview.com","license":"Apache-2.0","open_issues":2,"watchers":2,"ai_summary":"A tool for evaluating AI agents by snapshotting their behavior and detecting regressions through diffing tool calls and sequences. Supports multiple AI platforms including LangGraph, CrewAI, OpenAI, and Anthropic.","readme_excerpt":"## Quick Start\n\n> **OpenAI adapter migration:** OpenAI shut down the Assistants API on **August 26, 2026**.\n> The latest published EvalView release, **0.8.1**, still uses that API; the Responses API\n> migration is currently **unreleased source**. If you use `openai-assistants`, follow the\n> [migration guide](docs/OPENAI_MIGRATION.md) before running your tests. An `assistant_id`\n> alone cannot preserve your agent's configuration. Other adapters are unaffected.\n\n```bash\npip install evalview\n```\n\n```bash\nevalview snapshot    # Record your agent's current behavior as the baseline\nevalview check       # After any change, diff against the baseline\n```\n\nThat's the whole loop. `check` returns one of:\n\n```\n  ✓ login-flow        PASSED          behavior matches baseline\n  ⚠ refund-request    TOOLS_CHANGED   called a different tool, or in a different order\n  ✗ billing-dispute   REGRESSION      score dropped — output quality fell\n```\n\nIt diffs the **whole trajectory** — tool names, parameters, and order — not just the final string. The deterministic tool + sequence diff runs offline, with no API key. Add an LLM judge only when you want output-quality scoring.\n\nExecuting your agent can still incur backend API charges: `--no-judge` skips the\njudge, not those calls. Embedding-based semantic comparison is opt-in.\n\nNo agent yet? See it work in 30 seconds:\n\n```bash\nevalview demo\n```","github_created_at":"2025-11-17T22:21:38+00:00","created_at":"2026-07-11T23:14:36.301489+00:00","updated_at":"2026-09-20T04:30:17.988604+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"},{"slug":"evaluation-observability","name":"Evaluation & Observability","url":"https://www.graphcanon.com/categories/evaluation-observability","markdown_url":"https://www.graphcanon.com/categories/evaluation-observability.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/evaluation-observability"}],"tags":[{"slug":"agent-benchmark","name":"agent-benchmark"},{"slug":"agent-evaluation","name":"agent-evaluation"},{"slug":"agentic-ai","name":"agentic-ai"},{"slug":"ai-agents","name":"ai-agents"},{"slug":"anthropic","name":"anthropic"},{"slug":"autogen","name":"autogen"},{"slug":"cli","name":"cli"},{"slug":"crewai","name":"crewai"}],"trust":{"provenance":{"is_fork":false,"github_id":1098589658,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-18T08:22:22.929Z","maintenance":{"label":"Active","score":82,"methodology":"github_public_v1","releases_90d":1,"days_since_push":13,"last_release_at":"2026-07-26T19:53:19Z","stars_delta_30d":8,"open_issues_delta_30d":-1},"security_summary":{"status":"findings","scanner":"mcp_manifest@v1","low_count":0,"high_count":0,"last_scan_at":"2026-09-18T08:22:25.066Z","medium_count":1,"scan_profile":"mcp_manifest","critical_count":0}},"capability_facts":{"mcp":{"source":"repo_scan","observed_at":"2026-09-18T08:22:21.868Z","server_manifest":false},"scan":{"source":"repo_scan","observed_at":"2026-09-18T08:22:21.868Z"},"deploy":{"source":"dockerfile:Dockerfile","self_host":true,"observed_at":"2026-09-18T08:22:21.868Z","managed_saas":false},"has_cli":{"value":true,"source":"pyproject.toml:[project.scripts]","observed_at":"2026-09-18T08:22:21.868Z"},"languages":{"value":["python","javascript","typescript"],"source":"github.language+package.json+pyproject.toml","observed_at":"2026-09-18T08:22:21.868Z"},"has_docker":{"value":true,"source":"dockerfile:Dockerfile","observed_at":"2026-09-18T08:22:21.868Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-09-18T08:22:21.868Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you need to snapshot and diff the behavior of AI agents across multiple platforms, including LangGraph, CrewAI, OpenAI, and Anthropic.","If you are working with AI agents and require a tool that can detect regressions in continuous integration (CI) environments.","When you want to ensure that changes in your AI agent's behavior do not introduce unintended regressions, as eval-view provides a deterministic tool and sequence diff that runs offline without API key","If you are using the Responses API and need to migrate from the deprecated OpenAI Assistants API, as eval-view offers a migration guide for this specific scenario."],"when_not_to_use":["If you are working exclusively with AI platforms not supported by eval-view, such as those not listed among LangGraph, CrewAI, OpenAI, and Anthropic.","When you do not require regression testing or behavior snapshotting for your AI agents, as eval-view is specifically designed for these purposes.","If you are looking for a tool that does not involve backend API charges for executing your agent, as eval-view does not skip these charges even with the --no-judge flag.","If you need a tool that automatically handles the migration from the OpenAI Assistants API to the Responses API without manual intervention, as eval-view requires following a migration guide for this."],"source":"enrich:decision_facts","observed_at":"2026-09-18T11:14:52.986Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"Eval-view is a Python-based tool for regression testing of AI agents, supporting multiple platforms like LangGraph, CrewAI, OpenAI, and Anthropic. It snapshots AI behavior and detects regressions through diffing tool and"}]}}