feat: persist runner execution logs and serve history via API

[0vertake]

No description provided.

[superplanehq-integration]

👋 Commands for maintainers:

• /sp start - Start an ephemeral machine (takes ~30s)
• /sp stop - Stop a running machine (auto-executed on pr close)

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit fee5f5c. Configure here.}

[cursor]

Bash runner drops upstream payload

High Severity

Run Bash now creates broker tasks with only commands and no message_chain, while JS/Python runners still send upstream data that way. Bash scripts and docs rely on SUPERPLANE_PAYLOAD_FILE being populated from that chain, so workflows reading upstream context can fail or see empty payload after this change.

 

^{Reviewed by Cursor Bugbot for commit fee5f5c. Configure here.}

[superplane-gh-integration-9000]

Docs Impact Review

This PR adds a new public API endpoint for retrieving persisted runner execution logs, which is a meaningful new capability not mentioned anywhere in the current docs.

Suggested docs updates:

• docs.superplane.com/concepts/runners — add a new "Execution log history" section explaining that runner node logs (stdout lines, command start/end events, errors) are automatically persisted and can be retrieved via the API at GET /api/v1/canvases/{canvas_id}/executions/{execution_id}/logs. Document the limit and after_sequence cursor-pagination parameters, and note that this endpoint is only available for runner component nodes (shell commands, bash, JavaScript, Python).

Why: The PR introduces ListNodeExecutionLogs as a new gRPC/REST endpoint (proto + authorization rule + service handler), persists structured log records from all four runner component types, and defines four distinct log entry types (line, error, cmd_start, cmd_end). The Runners concept page currently has no mention of log persistence or log retrieval; users discovering the new endpoint in Swagger would have no conceptual documentation explaining what fields to expect or how pagination works.

---

Maintainers: Reply with /docs-agree to open a tracking issue in superplanehq/docs (and assign it to you), or /docs-reject to dismiss this check.

Posted automatically by warp-gateway · commit 0fac001

[AleksandarCole]

/docs-reject

[lucaspin]

Why are we storing these in the database? Shouldn't we use S3?