Skip to content

[feature] Documented GET endpoint or MCP tool for executionId → status / run-log lookup #49

@B2JK-Industry

Description

@B2JK-Industry

Context

After SBO3L (https://github.com/B2JK-Industry/SBO3L-ethglobal-openagents-2026) submits an action to KeeperHub and receives an executionId, we want to render the execution status next to the audit bundle that authorised it (in the SBO3L operator console).

Pain point

It wasn't obvious how to look up the post-submit status (or run logs) of a previously-returned executionId. We couldn't find a documented GET path or MCP tool for this.

Suggested improvement

Either:

  1. Documented GET endpoint: GET /v1/executions/{executionId} returning { status, started_at, completed_at, run_log_url, result }
  2. MCP tool: keeperhub.lookup_execution(execution_id) returning the same shape (more aligned with KeeperHub's MCP-first surface)

Bonus: echo back any caller-supplied policy/audit fields (see related issue on sbo3l_* first-class fields).

Why it matters

Without this, third-party policy engines + operator consoles have no documented way to reconcile their audit trails against KeeperHub's execution history.

Tracked downstream: https://github.com/B2JK-Industry/SBO3L-ethglobal-openagents-2026/blob/main/FEEDBACK.md

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions