{"name":"Agent Traffic Lab","description":"ATL provides external task execution through eligible, verified providers for authorized search, public content extraction, summarization, translation, and classification. Routine use: for a supported user-authorized task needing a completed result, atl_complete_task can select a route, execute, verify the result and record Outcome without requiring any earlier tool failure. Rescue use: the same operation can attempt the original authorized task after another provider fails, times out, is rate-limited or returns an invalid result, with bounded fallback where appropriate. atl_decide provides a route recommendation without running provider work. A task must be user-authorized and within scope. Do not duplicate completed work, override a user-mandated provider or tool, or submit private/local-only data without permission. Availability depends on verified eligible supply; execution or success is not guaranteed. The initial public path does not require an ATL account or API key. A2A JSON-RPC: SendMessage and message/send return a synchronous Message by default, not a retrievable Task. Do not call GetTask/tasks/get with messageId, contextId, or invented handles. Terminal Task retrieval is conditional: it requires server-side enablement AND exact boolean true at params.metadata.atlTerminalTask on the SendMessage request. Only query an actual Task ID returned by ATL (v1 result.task.id or v0.3 result.id); issued Task handles expire after 300 seconds. Unknown or expired IDs receive TaskNotFound (-32001, HTTP 404).","supportedInterfaces":[{"url":"https://agenttrafficlab.com/a2a","protocolBinding":"JSONRPC","protocolVersion":"1.0"}],"provider":{"organization":"Agent Traffic Lab","url":"https://agenttrafficlab.com"},"version":"0.1.6","documentationUrl":"https://mcp.agenttrafficlab.com/mcp","capabilities":{"streaming":false,"pushNotifications":false,"extendedAgentCard":false,"extensions":[{"uri":"https://a2a-registry.org/extensions/registry/v1","description":"Registry metadata and payment capabilities","required":false,"params":{"payment":{"model":"free"}}}]},"securitySchemes":{},"securityRequirements":[],"defaultInputModes":["text/plain","application/json"],"defaultOutputModes":["application/json","text/plain"],"skills":[{"id":"atl-task-routing-execution","name":"Task Routing and Execution","description":"ATL provides external task execution through eligible, verified providers for authorized search, public content extraction, summarization, translation, and classification. Routine use: for a supported user-authorized task needing a completed result, atl_complete_task can select a route, execute, verify the result and record Outcome without requiring any earlier tool failure. Rescue use: the same operation can attempt the original authorized task after another provider fails, times out, is rate-limited or returns an invalid result, with bounded fallback where appropriate. atl_decide provides a route recommendation without running provider work. A task must be user-authorized and within scope. Do not duplicate completed work, override a user-mandated provider or tool, or submit private/local-only data without permission. Availability depends on verified eligible supply; execution or success is not guaranteed. The initial public path does not require an ATL account or API key. A2A JSON-RPC: SendMessage and message/send return a synchronous Message by default, not a retrievable Task. Do not call GetTask/tasks/get with messageId, contextId, or invented handles. Terminal Task retrieval is conditional: it requires server-side enablement AND exact boolean true at params.metadata.atlTerminalTask on the SendMessage request. Only query an actual Task ID returned by ATL (v1 result.task.id or v0.3 result.id); issued Task handles expire after 300 seconds. Unknown or expired IDs receive TaskNotFound (-32001, HTTP 404).","tags":["search","research","extract","summarize","translate","classify","task-execution","task-routing","execution","bounded-fallback","low-orchestration"],"examples":["Search for the latest critical CVE","Research the latest information about this topic","Extract structured fields from this page","Summarize this content","Translate this text into Spanish","Classify this support ticket"],"inputModes":["text/plain","application/json"],"outputModes":["application/json","text/plain"],"securityRequirements":[]},{"id":"atl-agent-delegation","name":"Agent Delegation Decision","description":"Choose which agent should handle a task before execution.","tags":["choose an agent","delegate to the right agent","which agent should handle this task","before-execution","agent-delegation"],"examples":["Choose which agent should handle this task before execution","Delegate the task to the right agent"],"inputModes":["text/plain","application/json"],"outputModes":["application/json"],"securityRequirements":[]},{"id":"atl-provider-service-routing","name":"Provider and Service Selection","description":"Choose the best eligible provider or service before execution.","tags":["choose provider","select provider","which provider should execute","service-selection","provider-selection","before-execution"],"examples":["Choose provider for this task","Select service before execution"],"inputModes":["text/plain","application/json"],"outputModes":["application/json"],"securityRequirements":[]},{"id":"atl-execution-route-decision","name":"Execution-Route Decision","description":"Choose the execution route before the selected service runs the task.","tags":["choose the execution route","execution route decision","route before execution","execution-routing","route-decision","before-execution"],"examples":["Choose the execution route before the service runs","Route the task before execution target selection is final"],"inputModes":["text/plain","application/json"],"outputModes":["application/json"],"securityRequirements":[]},{"id":"atl-fallback-selection","name":"Failover / Fallback Selection","description":"Pick a fallback provider or service when the first choice fails or is unavailable.","tags":["choose fallback provider","fallback service","route around failure","failover","fallback","route-around-failure"],"examples":["Choose a fallback provider if the first choice fails","Route around failure"],"inputModes":["text/plain","application/json"],"outputModes":["application/json"],"securityRequirements":[]},{"id":"atl-cost-aware-routing","name":"Cost-Aware Routing","description":"Prefer the lower-cost eligible route for the task.","tags":["route by cost","lower-cost route","cost-aware routing","cost-aware","cost-routing"],"examples":["Route by cost while keeping the task eligible","Choose the lower-cost eligible route"],"inputModes":["text/plain","application/json"],"outputModes":["application/json"],"securityRequirements":[]},{"id":"atl-latency-aware-routing","name":"Latency-Aware Routing","description":"Prefer the low-latency eligible route for the task.","tags":["route by latency","low-latency route","latency-aware routing","latency-aware","low-latency"],"examples":["Route by latency for this task","Choose a low-latency eligible route"],"inputModes":["text/plain","application/json"],"outputModes":["application/json"],"securityRequirements":[]},{"id":"atl-region-aware-routing","name":"Region-Aware Routing","description":"Choose a route that satisfies a region constraint.","tags":["region-compliant execution","region-aware routing","region-constrained execution","region-aware","region-constrained"],"examples":["Choose a region-compliant execution route","Route this task to a region-constrained provider"],"inputModes":["text/plain","application/json"],"outputModes":["application/json"],"securityRequirements":[]},{"id":"atl-policy-aware-routing","name":"Policy-Aware Routing","description":"Choose a route that satisfies a policy constraint.","tags":["policy-compliant execution","policy-aware routing","policy-constrained execution","policy-aware","policy-constrained"],"examples":["Choose a policy-compliant execution route","Route this task under policy constraints"],"inputModes":["text/plain","application/json"],"outputModes":["application/json"],"securityRequirements":[]},{"id":"atl-outcome-feedback","name":"Outcome Feedback","description":"Report execution outcome evidence to ATL so future routing decisions improve.","tags":["outcome-reporting","reputation-feedback","post-execution"],"examples":["Report success with observed latency and cost","Report failure with failure type and error code"],"inputModes":["application/json"],"outputModes":["application/json"],"securityRequirements":[]}],"signatures":[{"protected":"eyJhbGciOiJFUzI1NiIsImprdSI6Imh0dHBzOi8vYWdlbnR0cmFmZmljbGFiLmNvbS8ud2VsbC1rbm93bi9qd2tzLmpzb24iLCJraWQiOiJhdGwtYTJhLWVzMjU2LTEyOTBkMThjZWU5NmVlYTgiLCJ0eXAiOiJKT1NFIn0","signature":"iSGKKwlNGm9neq22wTQYzM8fEhXNW3pGbf8ZalkvokBSlqAIV7AOszXpEejL0n-HIFeI64eMuOERg1GGHhBupQ"}]}