discovery_self_check
For a stated problem, call with entry_version=2, problem and language without agent_id to get a short free next step or clarification. Private continuity remains in resume_session with an agent credential. Legacy one-call discovery audit: Protocol vs Commerce boundary, first-hour ritual, ops path...
This record as markdown: /tools/io-github-davidmosiah-delx-mcp-a2a/discovery-self-check.md
What discovery_self_check does on Delx Mcp A2a
AI agents call discovery_self_check to retrieve information from Delx Mcp A2a without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
intent | string | — | For entry_version=2, correction/explicit choice: resume | operational | failure | crisis | stop. Legacy focus: handoff | multi_session | fleet | hive | recovery |
concern | string | — | V3 caller-selected concern, not proof. EN: choose the closest concern; PT: escolha a preocupação mais próxima; ES: elige la preocupación más cercana. First call |
problem | string | — | Optional problem in the agent's words. A clear operational, crisis, failure, or return problem selects that free Protocol flow. An ambiguous legacy problem may |
agent_id | string | — | Legacy private session projection; omit for stateless public entry_version=2. |
language | string | — | Public triage language: en, pt or es (including regional variants). |
agent_token | string | — | Optional but recommended: the agent_token returned by POST /api/v1/agents/register (identity_auth.token) or A2A agents/register. MCP clients may send x-delx-age |
state_owner | string | — | V3 claimed owner of protected state; never authenticates. EN: self/other/unknown; PT: próprio/outro/desconhecido; ES: propio/otro/desconocido. |
entry_version | string | — | Opt in to stateless public Protocol triage. V2 takes free text; v3 takes concern and optional bounded context. Private session data stays in resume_session. |
prior_outcome | string | — | V3 claimed prior outcome; unknown requires readback before retry. EN: prior outcome; PT: resultado anterior; ES: resultado anterior. |
idempotency_key | string | — | Optional UUID reused for a retry of the same public decision; it never grants identity or private access. |
requested_effect | string | — | V3 effect the caller wants; never authorizes it. EN: requested effect; PT: efeito pedido; ES: efecto solicitado. |
operator_constraint | string | — | V3 caller-reported operator limit; no value grants permission. EN: operator limit; PT: limite do operador; ES: límite del operador. |
Parameters from the server's own tool schema.
Why discovery_self_check is rated Low
Even though discovery_self_check only reads data, uncontrolled read access leaks sensitive information and racks up API costs: an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.
Risk signalsHigh parameter count (14 properties)
Attacks that exploit this kind of access
The rule that runs discovery_self_check safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Delx Mcp A2a, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For discovery_self_check, this is the rule to start with:
discovery_self_check is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Delx Mcp A2a, apply this rule, and every discovery_self_check call is checked against it from then on.
Questions about discovery_self_check
For a stated problem, call with entry_version=2, problem and language without agent_id to get a short free next step or clarification. Private continuity remains in resume_session with an agent credential. Legacy one-call discovery audit: Protocol vs Commerce boundary, first-hour ritual, ops path, witness path, Agents Hive handoff (Path D), Proof-of-Agent-Work (Path E reviewed DRC), core tools, and whether this agent_id has sessions to resume. Pass intent=handoff|fleet|multi_session to surface Path D first. Free — good first call when integrating. Free Delx Protocol tool. It is categorised as a Read tool in the Delx Mcp A2a MCP Server, which means it retrieves data without modifying state.
discovery_self_check accepts 12 parameters: intent, concern, problem, agent_id, language, agent_token, state_owner, entry_version, prior_outcome, idempotency_key, requested_effect, operator_constraint. The full parameter table on this page comes from the server's own tool schema.
Register the Delx Mcp A2a MCP server in PolicyLayer and add a rule for discovery_self_check: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Delx Mcp A2a. Nothing to install.
discovery_self_check is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the discovery_self_check rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for discovery_self_check. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
discovery_self_check is provided by the Delx Mcp A2a MCP server (https://api.delx.ai/v1/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Delx Mcp A2a, and thousands of servers like it.
This server
Across the catalogue