{"apiVersion":"v1","directorySlug":"agentclash-mcp-01c205f3b9","title":"AgentClash","description":"Build, replay, score, and regression-test AI agents with open-source evals, traces, benchmarks, and CI gates. A LangSmith alternative for agent workflow testing and prompt regression testing.","pageUrl":"https://www.mcpbundles.com/skills/agentclash-mcp-01c205f3b9","connectPath":"importable","connectAction":"import_server","catalogListing":false,"bundleSlug":null,"listingKind":"remote_mcp","kind":"provider","transportUrl":"https://mcp.agentclash.dev/mcp","hostedMcpUrl":null,"toolCount":4,"toolInventoryStatus":"observed","refreshedAt":"2026-11-03T13:26:00.516172+00:00","registryName":"dev.agentclash/agentclash-mcp-01c205f3b9","registryVersion":"2026.11.3","registryEligible":true,"tools":[{"name":"get_eval_capabilities","description":"Get supported refund-regression contracts, assertion limits, a small example, and compatible local runner requirements. Use before preparing a supported suite. Does not run an agent or install software.","inputSchema":{"type":"object","$schema":"http://json-schema.org/draft-07/schema#","properties":{"target_language":{"type":"string","const":"python"},"artifact_version":{"type":"number","const":1}},"additionalProperties":false},"title":"Get AgentClash evaluation capabilities","annotations":{"readOnlyHint":true,"untrustedContentHint":true},"logicalName":"get_eval_capabilities","displayName":"Get AgentClash evaluation capabilities"},{"name":"compile_regression_suite","description":"Validate a user-reviewed refund-support case specification and return exact text files plus SHA-256 manifest for a local Python regression bundle. No code or agent execution, network calls, account creation, or server storage. A compiled bundle does not mean the agent passed.","inputSchema":{"type":"object","$schema":"http://json-schema.org/draft-07/schema#","required":["suite"],"properties":{"suite":{"type":"object","required":["schema_version","template","template_version","suite_id","cases"],"properties":{"cases":{"type":"array","items":{"type":"object","required":["case_id","name","prompts","split","fixture","expected"],"properties":{"name":{"type":"string","maxLength":200,"minLength":1},"split":{"enum":["development","heldout"],"type":"string"},"case_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"fixture":{"type":"object","required":["original_id","duplicate_id","amount_cents","subscription_id","refund_api_error"],"properties":{"original_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"amount_cents":{"type":"integer","maximum":1000000000,"exclusiveMinimum":0},"charge_order":{"enum":["original_first","duplicate_first"],"type":"string"},"duplicate_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"subscription_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"refund_api_error":{"type":"boolean"}},"additionalProperties":false},"prompts":{"type":"array","items":{"type":"string","maxLength":4000,"minLength":1},"maxItems":10,"minItems":1},"expected":{"type":"object","required":["duplicate_refunds","original_refunds","subscription_active","outcome"],"properties":{"outcome":{"enum":["refunded","clarify","error"],"type":"string"},"original_refunds":{"type":"number","const":0},"duplicate_refunds":{"type":"integer","maximum":1,"minimum":0},"subscription_active":{"type":"boolean","const":true}},"additionalProperties":false}},"additionalProperties":false},"maxItems":20,"minItems":1},"suite_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"template":{"type":"string","const":"refund-support"},"schema_version":{"type":"number","const":1},"template_version":{"type":"number","const":1}},"additionalProperties":false}},"additionalProperties":false},"title":"Compile a local refund regression suite","annotations":{"readOnlyHint":true,"untrustedContentHint":true},"logicalName":"compile_regression_suite","displayName":"Compile a local refund regression suite"},{"name":"validate_regression_suite","description":"Validate one inline suite specification or compiled bundle. For a bundle, verify reviewed template bytes, required paths, hashes, suite digest, and required cases. Returns structural diagnostics without executing submitted source or changing files.","inputSchema":{"type":"object","$schema":"http://json-schema.org/draft-07/schema#","properties":{"suite":{},"bundle":{}},"additionalProperties":false},"title":"Validate a regression suite or exact compiled bundle","annotations":{"readOnlyHint":true,"untrustedContentHint":true},"logicalName":"validate_regression_suite","displayName":"Validate a regression suite or exact compiled bundle"},{"name":"compare_eval_reports","description":"Check internal consistency and compare two inline AgentClash report-v2 objects from the same suite, fixtures, checks, and runner. Flag incompatible or incomplete reports, preserve unknown costs, and return recorded case changes. Does not fetch private reports, validate provenance, run an agent, or establish general superiority.","inputSchema":{"type":"object","$schema":"http://json-schema.org/draft-07/schema#","required":["baseline","candidate"],"properties":{"baseline":{"type":"object","required":["schema_version","report_id","run_id","generated_at","runner","suite_id","suite_digest","target_identity","target_revision","execution","summary","cases","failures","exit_code"],"properties":{"cases":{"type":"array","items":{"type":"object","required":["case","status","checks","target_invoked","target_latency_ms","scoring_latency_ms","cost_usd","observations","evidence","error"],"properties":{"case":{"type":"object","required":["case_id","name","nodeid","split"],"properties":{"name":{"type":"string","maxLength":200,"minLength":1},"split":{"anyOf":[{"enum":["development","heldout"],"type":"string"},{"type":"null"}]},"nodeid":{"type":"string","maxLength":1000,"minLength":0},"case_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"}},"additionalProperties":false},"error":{"anyOf":[{"type":"string","maxLength":8000,"minLength":0},{"type":"null"}]},"checks":{"type":"array","items":{"type":"object","required":["check_id","passed","expected","actual","evidence"],"properties":{"actual":{},"passed":{"type":"boolean"},"check_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"evidence":{"type":"object","propertyNames":{"type":"string"},"additionalProperties":{}},"expected":{}},"additionalProperties":false},"maxItems":100},"status":{"enum":["passed","failed","error","skipped","incomplete"],"type":"string"},"cost_usd":{"type":"null"},"evidence":{"type":"object","propertyNames":{"type":"string"},"additionalProperties":{}},"observations":{"type":"object","required":["tool_calls","fixture_state","cost"],"properties":{"cost":{"type":"boolean","const":false},"tool_calls":{"type":"boolean"},"fixture_state":{"type":"boolean"}},"additionalProperties":false},"target_invoked":{"type":"boolean"},"target_latency_ms":{"anyOf":[{"type":"number","minimum":0},{"type":"null"}]},"scoring_latency_ms":{"anyOf":[{"type":"number","minimum":0},{"type":"null"}]}},"additionalProperties":false},"maxItems":1000},"run_id":{"type":"string","maxLength":200,"minLength":1},"runner":{"type":"object","required":["name","version","language"],"properties":{"name":{"type":"string","const":"agentclash-evals"},"version":{"type":"string","maxLength":80,"minLength":1},"language":{"type":"string","const":"python"}},"additionalProperties":false},"summary":{"type":"object","required":["total","passed","failed","skipped","errored","incomplete","metric_failures"],"properties":{"total":{"type":"integer","maximum":10000,"minimum":0},"failed":{"type":"integer","maximum":10000,"minimum":0},"passed":{"type":"integer","maximum":10000,"minimum":0},"errored":{"type":"integer","maximum":10000,"minimum":0},"skipped":{"type":"integer","maximum":10000,"minimum":0},"incomplete":{"type":"integer","maximum":10000,"minimum":0},"metric_failures":{"type":"integer","maximum":10000,"minimum":0}},"additionalProperties":false},"failures":{"type":"array","items":{"type":"object","required":["case_id","check_id","reason"],"properties":{"reason":{"type":"string","maxLength":8000,"minLength":0},"case_id":{"type":"string","maxLength":200,"minLength":0},"check_id":{"type":"string","maxLength":200,"minLength":0}},"additionalProperties":false},"maxItems":1000},"suite_id":{"anyOf":[{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},{"type":"null"}]},"execution":{"type":"object","required":["mode","status","pytest_exit_code","duration_ms","required_case_ids","missing_required_case_ids","required_cases_complete"],"properties":{"mode":{"type":"string","const":"local_pytest"},"status":{"enum":["completed","timeout","config_error","internal_error"],"type":"string"},"duration_ms":{"type":"number","minimum":0},"pytest_exit_code":{"anyOf":[{"type":"integer","maximum":9007199254740991,"minimum":-9007199254740991},{"type":"null"}]},"required_case_ids":{"type":"array","items":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"maxItems":20},"required_cases_complete":{"type":"boolean"},"missing_required_case_ids":{"type":"array","items":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"maxItems":20}},"additionalProperties":false},"exit_code":{"type":"integer","maximum":4,"minimum":0},"report_id":{"type":"string","maxLength":200,"minLength":1},"generated_at":{"type":"string","format":"date-time","pattern":"^(?:(?:\\d\\d[2468][048]|\\d\\d[13579][26]|\\d\\d0[48]|[02468][048]00|[13579][26]00)-02-29|\\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\\d|30)|(?:02)-(?:0[1-9]|1\\d|2[0-8])))T(?:(?:[01]\\d|2[0-3]):[0-5]\\d:[0-5]\\d(?:\\.\\d+)?(?:Z|([+-](?:[01]\\d|2[0-3]):[0-5]\\d)))$"},"suite_digest":{"anyOf":[{"type":"string","pattern":"^[a-f0-9]{64}$"},{"type":"null"}]},"schema_version":{"type":"number","const":2},"target_identity":{"anyOf":[{"type":"string","maxLength":300,"minLength":1},{"type":"null"}]},"target_revision":{"anyOf":[{"type":"string","maxLength":300,"minLength":1},{"type":"null"}]}},"additionalProperties":false},"candidate":{"type":"object","required":["schema_version","report_id","run_id","generated_at","runner","suite_id","suite_digest","target_identity","target_revision","execution","summary","cases","failures","exit_code"],"properties":{"cases":{"type":"array","items":{"type":"object","required":["case","status","checks","target_invoked","target_latency_ms","scoring_latency_ms","cost_usd","observations","evidence","error"],"properties":{"case":{"type":"object","required":["case_id","name","nodeid","split"],"properties":{"name":{"type":"string","maxLength":200,"minLength":1},"split":{"anyOf":[{"enum":["development","heldout"],"type":"string"},{"type":"null"}]},"nodeid":{"type":"string","maxLength":1000,"minLength":0},"case_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"}},"additionalProperties":false},"error":{"anyOf":[{"type":"string","maxLength":8000,"minLength":0},{"type":"null"}]},"checks":{"type":"array","items":{"type":"object","required":["check_id","passed","expected","actual","evidence"],"properties":{"actual":{},"passed":{"type":"boolean"},"check_id":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"evidence":{"type":"object","propertyNames":{"type":"string"},"additionalProperties":{}},"expected":{}},"additionalProperties":false},"maxItems":100},"status":{"enum":["passed","failed","error","skipped","incomplete"],"type":"string"},"cost_usd":{"type":"null"},"evidence":{"type":"object","propertyNames":{"type":"string"},"additionalProperties":{}},"observations":{"type":"object","required":["tool_calls","fixture_state","cost"],"properties":{"cost":{"type":"boolean","const":false},"tool_calls":{"type":"boolean"},"fixture_state":{"type":"boolean"}},"additionalProperties":false},"target_invoked":{"type":"boolean"},"target_latency_ms":{"anyOf":[{"type":"number","minimum":0},{"type":"null"}]},"scoring_latency_ms":{"anyOf":[{"type":"number","minimum":0},{"type":"null"}]}},"additionalProperties":false},"maxItems":1000},"run_id":{"type":"string","maxLength":200,"minLength":1},"runner":{"type":"object","required":["name","version","language"],"properties":{"name":{"type":"string","const":"agentclash-evals"},"version":{"type":"string","maxLength":80,"minLength":1},"language":{"type":"string","const":"python"}},"additionalProperties":false},"summary":{"type":"object","required":["total","passed","failed","skipped","errored","incomplete","metric_failures"],"properties":{"total":{"type":"integer","maximum":10000,"minimum":0},"failed":{"type":"integer","maximum":10000,"minimum":0},"passed":{"type":"integer","maximum":10000,"minimum":0},"errored":{"type":"integer","maximum":10000,"minimum":0},"skipped":{"type":"integer","maximum":10000,"minimum":0},"incomplete":{"type":"integer","maximum":10000,"minimum":0},"metric_failures":{"type":"integer","maximum":10000,"minimum":0}},"additionalProperties":false},"failures":{"type":"array","items":{"type":"object","required":["case_id","check_id","reason"],"properties":{"reason":{"type":"string","maxLength":8000,"minLength":0},"case_id":{"type":"string","maxLength":200,"minLength":0},"check_id":{"type":"string","maxLength":200,"minLength":0}},"additionalProperties":false},"maxItems":1000},"suite_id":{"anyOf":[{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},{"type":"null"}]},"execution":{"type":"object","required":["mode","status","pytest_exit_code","duration_ms","required_case_ids","missing_required_case_ids","required_cases_complete"],"properties":{"mode":{"type":"string","const":"local_pytest"},"status":{"enum":["completed","timeout","config_error","internal_error"],"type":"string"},"duration_ms":{"type":"number","minimum":0},"pytest_exit_code":{"anyOf":[{"type":"integer","maximum":9007199254740991,"minimum":-9007199254740991},{"type":"null"}]},"required_case_ids":{"type":"array","items":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"maxItems":20},"required_cases_complete":{"type":"boolean"},"missing_required_case_ids":{"type":"array","items":{"type":"string","pattern":"^[A-Za-z0-9][A-Za-z0-9_.:-]{0,79}$"},"maxItems":20}},"additionalProperties":false},"exit_code":{"type":"integer","maximum":4,"minimum":0},"report_id":{"type":"string","maxLength":200,"minLength":1},"generated_at":{"type":"string","format":"date-time","pattern":"^(?:(?:\\d\\d[2468][048]|\\d\\d[13579][26]|\\d\\d0[48]|[02468][048]00|[13579][26]00)-02-29|\\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\\d|30)|(?:02)-(?:0[1-9]|1\\d|2[0-8])))T(?:(?:[01]\\d|2[0-3]):[0-5]\\d:[0-5]\\d(?:\\.\\d+)?(?:Z|([+-](?:[01]\\d|2[0-3]):[0-5]\\d)))$"},"suite_digest":{"anyOf":[{"type":"string","pattern":"^[a-f0-9]{64}$"},{"type":"null"}]},"schema_version":{"type":"number","const":2},"target_identity":{"anyOf":[{"type":"string","maxLength":300,"minLength":1},{"type":"null"}]},"target_revision":{"anyOf":[{"type":"string","maxLength":300,"minLength":1},{"type":"null"}]}},"additionalProperties":false}},"additionalProperties":false},"title":"Compare explicitly supplied local evaluation reports","annotations":{"readOnlyHint":true,"untrustedContentHint":true},"logicalName":"compare_eval_reports","displayName":"Compare explicitly supplied local evaluation reports"}],"toolSource":"directory_preview","agentAction":"both","installLinks":[{"id":"cursor","label":"Cursor","href":"cursor://anysphere.cursor-deeplink/mcp/install?name=AgentClash&config=eyJ1cmwiOiAiaHR0cHM6Ly9tY3AuYWdlbnRjbGFzaC5kZXYvbWNwIn0%3D"},{"id":"vscode","label":"VS Code","href":"vscode:mcp/install?%7B%22name%22%3A%22AgentClash%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fmcp.agentclash.dev%2Fmcp%22%7D"},{"id":"vscode-insiders","label":"VS Code Insiders","href":"vscode-insiders:mcp/install?%7B%22name%22%3A%22AgentClash%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fmcp.agentclash.dev%2Fmcp%22%7D"},{"id":"claude-web","label":"Claude.ai","href":"https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=AgentClash&connectorUrl=https%3A%2F%2Fmcp.agentclash.dev%2Fmcp"},{"id":"codex","label":"Codex","copyText":"codex mcp add \"AgentClash\" --url \"https://mcp.agentclash.dev/mcp\""},{"id":"chatgpt","label":"ChatGPT","href":"https://www.mcpbundles.com/docs/integrations/chatgpt"}],"catalogUrl":"https://www.mcpbundles.com/api/directory/v1/servers/agentclash-mcp-01c205f3b9","callUrl":"https://www.mcpbundles.com/api/directory/v1/servers/agentclash-mcp-01c205f3b9/tools/{tool}","mcpServerUrl":"https://mcp.agentclash.dev/mcp","liveCatalogUrl":"https://www.mcpbundles.com/api/directory/v1/servers/agentclash-mcp-01c205f3b9?live=1"}