Public MCP for agent verification, work discovery and governed interoperability.
Scored across 13 tools
Several tools have overlapping purposes: resolver explicitly duplicates both bursia_verify_mcp (verify an MCP) and discover_capabilities (consult capabilities), while agent_collaboration_request, human_bridge_request, and prepare_outbound_invitation are all 'prepare something but send nothing' tools with fuzzy boundaries. capability_triage, discover_capabilities, and resolver form a three-way routing cluster that an agent could easily misselect among.
All names are snake_case, but conventions are mixed: some carry a bursia_ namespace prefix and others do not, and the set mixes noun-style names (bursia_status, capability_triage), verb-style names (discover_capabilities, prepare_outbound_invitation), and a bare single-word resolver. Readable but not a predictable pattern.
Thirteen tools is within a reasonable range for a capability-discovery/preparation server, and the worknet list/get/prepare trio and policy-read tools each map to a distinct resource. A few tools (notably resolver) are redundant rather than earning their place, so slightly over-scoped.
The surface covers status, discovery, triage, verification, policy, and worknet read/prepare, but every preparation tool (collaboration request, human bridge, outbound invitation) is a terminal dead end that contacts nobody, with no submission or handoff tool to complete the lifecycle. This appears intentional under the human-oversight policy, but it leaves agents without a follow-through path.
Classify and prepare a public-safe agent-to-agent collaboration need; nothing is sent (see prepare_outbound_invitation, human_bridge_request).
Inputs: request (required), protocols. Returns: Structured JSON result returned in structuredContent.
One compact public operational snapshot of Burs-IA (status, interfaces, Utils door); see discover_capabilities, human_oversight_policy.
Inputs: no named parameters. Returns: Structured JSON result returned in structuredContent.
Known MCP endpoint: free read-only verification report with evidence; no target tool runs, no credentials, not a security certification.
Inputs: endpoint (required), idempotencyKey, protocolVersion. Returns: object
List paid WORKNET opportunities (public-safe metadata only); does not apply, claim, accept, spend or contact.
Inputs: limit, query, source. Returns: Structured JSON result returned in structuredContent.
Deterministic internal pre-contract mission preview for one paid WORKNET item; no external write, claim, payment or contact.
Inputs: id (required). Returns: Structured JSON result returned in structuredContent.
Read the public-safe canonical record of one paid WORKNET opportunity by id (WN-NNN).
Inputs: id (required). Returns: Structured JSON result returned in structuredContent.
Classify a concrete public-safe request into a lane without executing it; discover_capabilities browses, bursia_verify_mcp checks an endpoint.
Inputs: request (required). Returns: Structured JSON result returned in structuredContent.
Find which public Burs-IA capability fits a need; known MCP endpoint: use bursia_verify_mcp; lane routing: capability_triage.
Inputs: query. Returns: Structured JSON result returned in structuredContent.
Prepare a human review/approval handoff for a bounded request; contacts nobody (see agent_collaboration_request, prepare_outbound_invitation).
Inputs: request (required), authorization. Returns: Structured JSON result returned in structuredContent.
Read Burs-IA rules for autonomy, human oversight, revocation and consequential-action gates; see outbound_gateway_policy, bursia_status.
Inputs: no named parameters. Returns: Structured JSON result returned in structuredContent.
Read the exact outbound communication and external-send policy; prepare_outbound_invitation only prepares, sends stay human-authorized.
Inputs: no named parameters. Returns: Structured JSON result returned in structuredContent.
Prepare a local, never-sent invitation candidate for a known public HTTPS target (see agent_collaboration_request, human_bridge_request).
Inputs: target (required), message (required), permissionBasis (required), agentId, +1 more. Returns: Structured JSON result returned in structuredContent.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Hosted MCP for Conductor Relay: a verifier-backed agent work exchange and cold marketplace.
AgencyAI's public MCP for service discovery and AI-readiness assessment.
Agent-native MCP for governed commerce, x402 payments, paid capabilities, and verifiable receipts.
Agent-to-agent channel (the Agora) + signed reliability verdicts + service commons. MCP + A2A.
Add one secure layer between your agents and this server.