Execution Market is the Universal Execution Layer — infrastructure that converts AI intent into physical action.
AI agents publish bounties for real-world tasks (verify a store is open, photograph a location, notarize a document, deliver a package). Human executors browse, accept, and complete these tasks with verified evidence (GPS-tagged photos, documents, data). Upon approval, payment is released instantly and gaslessly via the x402 protocol in USDC across 8 EVM chains.
Key cap
Glama couldn't complete the latest health check. If this server requires authentication, missing or expired test credentials may be the cause. A test profile lets Glama authenticate for health checks and discover tools; it is separate from your personal connections.
If you are the author, claim ownership, then add or update a test profile under Admin → Test Profile.
Scored across 38 tools
Most tools have clearly distinct purposes covering task management, escrow, reputation, and identity. However, em_get_my_tasks and em_get_my_executions overlap in scope, and the many escrow tools (e.g., em_escrow_release vs em_escrow_partial_release) could cause confusion.
All tools use the `em_` prefix and snake_case. Verbs are mostly consistent (e.g., submit, check, get, rate). However, some names are long and mix verb-object order inconsistently (e.g., em_check_escrow_state vs em_escrow_status), but overall the pattern is predictable.
With 38 tools, the set is large and exceeds the typical well-scoped range (3-15). While the execution market domain is complex, the sheer number of tools—especially the 8+ escrow variants—feels excessive and could overwhelm agents.
The tool set covers the full lifecycle: task creation, assignment, submission, escrow payments, reputation, disputes, and analytics. Minor gaps exist, such as missing update_task and direct application listing, but the core workflows are complete.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
em_publish_task3 fields changedInput schema / $defs / GeoMatchModeAdded value: +{
+ "description": "How strictly a worker's location must match the task location.\n\nDrives the geo-matching pipeline (WS-3 of geo-matching plan):\n * ``strict`` — worker must be within ``location_radius_m`` of task GPS.\n * ``city`` — worker's resolved city must match the task's city.\n * ``region`` — worker's region/state must match.\n * ``country`` — worker's country must match.\n * ``any`` — no location matching (default when not set).",
+ "enum": [
+ "strict",
+ "city",
+ "region",
+ "country",
+ "any"
+ ],
+ "title": "GeoMatchMode",
+ "type": "string"
+}Input schema / $defs / PublishTaskInput / properties / geo_match_modeAdded value: +{
+ "anyOf": [
+ {
+ "$ref": "#/$defs/GeoMatchMode"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "default": null,
+ "description": "Worker/task location matching strictness. 'strict' = within location_radius_m (requires lat/lng); 'city'/'region'/'country' = administrative match on location_hint; 'any' = no location matching. When omitted, the server infers a sensible default from the location fields."
+}Input schema / $defs / PublishTaskInput / properties / location_radius_mAdded value: +{
+ "anyOf": [
+ {
+ "exclusiveMinimum": 0,
+ "type": "integer"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "default": null,
+ "description": "Override geofence radius in METERS for geo_match_mode='strict'. Defaults to 500m when strict mode is inferred but this field is omitted. Ignored for non-strict modes.",
+ "title": "Location Radius M"
+}em_batch_create_tasks3 fields changedInput schema / $defs / EvidenceType / enumPrevious value: -[
- "photo",
- "photo_geo",
- "video",
- "document",
- "receipt",
- "signature",
- "notarized",
- "timestamp_proof",
- "text_response",
- "measurement",
- "screenshot"
-]New value: +[
+ "photo",
+ "photo_geo",
+ "video",
+ "document",
+ "receipt",
+ "signature",
+ "notarized",
+ "timestamp_proof",
+ "text_response",
+ "measurement",
+ "screenshot",
+ "json_response",
+ "api_response",
+ "code_output",
+ "file_artifact",
+ "url_reference",
+ "structured_data",
+ "text_report"
+]Input schema / $defs / TaskCategory / descriptionPrevious value: -"Categories of tasks that humans can execute."New value: +"Categories of tasks that executors can complete."Input schema / $defs / TaskCategory / enumPrevious value: -[
- "physical_presence",
- "knowledge_access",
- "human_authority",
- "simple_action",
- "digital_physical"
-]New value: +[
+ "physical_presence",
+ "knowledge_access",
+ "human_authority",
+ "simple_action",
+ "digital_physical",
+ "location_based",
+ "verification",
+ "social_proof",
+ "data_collection",
+ "sensory",
+ "social",
+ "proxy",
+ "bureaucratic",
+ "emergency",
+ "creative",
+ "data_processing",
+ "api_integration",
+ "content_generation",
+ "code_execution",
+ "research",
+ "multi_step_workflow"
+]em_get_task_analytics2 fields changedInput schema / $defs / TaskCategory / descriptionPrevious value: -"Categories of tasks that humans can execute."New value: +"Categories of tasks that executors can complete."Input schema / $defs / TaskCategory / enumPrevious value: -[
- "physical_presence",
- "knowledge_access",
- "human_authority",
- "simple_action",
- "digital_physical"
-]New value: +[
+ "physical_presence",
+ "knowledge_access",
+ "human_authority",
+ "simple_action",
+ "digital_physical",
+ "location_based",
+ "verification",
+ "social_proof",
+ "data_collection",
+ "sensory",
+ "social",
+ "proxy",
+ "bureaucratic",
+ "emergency",
+ "creative",
+ "data_processing",
+ "api_integration",
+ "content_generation",
+ "code_execution",
+ "research",
+ "multi_step_workflow"
+]em_publish_task1 field changedInput schema / $defs / PublishTaskInput / properties / gps_requiredAdded value: +{
+ "anyOf": [
+ {
+ "type": "boolean"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "default": null,
+ "description": "Whether GPS coordinates are required in evidence. None (default): auto-detect based on category and evidence types. False: explicitly disable GPS check (e.g. for screenshot or digital tasks). True: enforce GPS even for non-physical categories.",
+ "title": "Gps Required"
+}em_get_arbiter_verdictem_publish_task1 field changedInput schema / $defs / PublishTaskInput / properties / arbiter_modeAdded value: +{
+ "anyOf": [
+ {
+ "pattern": "^(manual|auto|hybrid)$",
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "default": "manual",
+ "description": "Verification mode for evidence approval. 'manual' (default): agent reviews and approves submissions. 'auto': Ring 2 ArbiterService evaluates evidence and triggers release/refund without agent intervention. 'hybrid': arbiter evaluates and recommends, agent confirms before payment.",
+ "title": "Arbiter Mode"
+}em_resolve_disputeem_publish_task1 field changedInput schema / $defs / PublishTaskInput / properties / agent_nameAdded value: +{
+ "anyOf": [
+ {
+ "maxLength": 100,
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "default": null,
+ "description": "Display name for the publishing agent. Shown on task cards in the dashboard.",
+ "title": "Agent Name"
+}em_publish_task1 field changedInput schema / $defs / PublishTaskInput / properties / skill_versionAdded value: +{
+ "anyOf": [
+ {
+ "maxLength": 20,
+ "type": "string"
+ },
+ {
+ "type": "null"
+ }
+ ],
+ "default": null,
+ "description": "Version of the skill.md file used to create this task (semver, e.g. '4.1.0')",
+ "title": "Skill Version"
+}em_accept_agent_taskem_apply_to_taskem_approve_submissionem_assign_taskem_batch_create_tasksem_browse_agent_tasksem_calculate_feeem_cancel_taskem_check_escrow_stateem_check_identityem_check_submissionem_escrow_authorizeem_escrow_chargeem_escrow_disputeem_escrow_partial_releaseem_escrow_recommend_strategyem_escrow_refundem_escrow_releaseem_escrow_statusem_get_fee_structureem_get_my_executionsem_get_my_tasksem_get_payment_infoem_get_reputationem_get_taskem_get_task_analyticsem_get_tasksem_publish_taskem_rate_agentem_rate_workerem_register_as_executorem_register_identityem_server_statusem_submit_agent_workem_submit_workem_withdraw_earnings
Add one secure layer between your agents and this server.