Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Assistant: claude-code Assistant-Model: opus Assistant-Process: 3775@bnt-lap001 Assistant-Session: cfacb6a5-ad5b-4f79-ac67-053e3b0d34a3
6.5 KiB
| id | type | title | domain | repo | status | flavor | owner | topic_slug | created | updated | depends_on | state_hub_workstream_id | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| RBT-WP-0006 | workplan | Execution mode and multi-repo pack index | infotech | runbook-tutorials | blocked | implementation | claude | stack | 2026-09-29 | 2026-10-03 |
|
35196f02-7cb9-5706-9529-1cba9c736058 |
Goal
Backlog (2026-09-29). Later capabilities, deliberately not started, each waiting for a decision from Bernd: opt-in execution of steps (auto-pilot),
discovery of packs across the estate, a runbook-aware feedback assistant, dependencies between runbooks, a terminal walk-through, the v0.3 design
that closes the gaps in docs/console-mapping.md, and an optional security and code review. Nothing here blocks the finished runbook-tutorials workplans.
Opt-in, audited step execution
id: RBT-WP-0006-T01
status: wait
needs_human: true
decision_id: 1e517449-2114-4a8f-9c6b-fcdbe977378e
blocking_reason: "Needs a decision on which identity runs the command."
priority: low
state_hub_task_id: "0424deab-3b51-5222-80b1-54c0ea26928d"
Fleet constraint (the-custodian standing notice, 2026-09-28): agent autonomy is granted per agent and scope, supervised first; promotion needs an explicit grant and
enforced EUR cost and risk limits, and a record or an acceptance rate does not authorize autopilot. Any execution mode here must be gated by such a grant.
Per-step execute: true allowed only for read-only steps first, with an
explicit operator confirmation, timeouts, and an audit record. Needs a
decision on which identity runs the command.
2026-10-03: human gate routed to hub decision 1e517449-2114-4a8f-9c6b-fcdbe977378e (D9: runbook-tutorials execution-mode decisions (RBT-WP-0006 T01–T07)) by CUST-WP-0075; resolving that decision raises C-40 here.
Cross-repo pack index
id: RBT-WP-0006-T02
status: wait
needs_human: true
decision_id: 1e517449-2114-4a8f-9c6b-fcdbe977378e
blocking_reason: "Needs a decision on scan scope."
priority: low
state_hub_task_id: "0cfb69bc-dc7e-5960-8359-64c49a742a6e"
rtut list --estate scans registered repos for runbooks/*/pack.yaml and publishes
a capability entry in the registry. Needs a decision on scan scope.
2026-10-03: human gate routed to hub decision 1e517449-2114-4a8f-9c6b-fcdbe977378e (D9: runbook-tutorials execution-mode decisions (RBT-WP-0006 T01–T07)) by CUST-WP-0075; resolving that decision raises C-40 here.
Runbook-aware feedback assistant
id: RBT-WP-0006-T03
status: wait
needs_human: true
decision_id: 1e517449-2114-4a8f-9c6b-fcdbe977378e
blocking_reason: "Needs decisions on access, identity and what is sent where."
priority: low
state_hub_task_id: "ce82d1ae-76a3-59c8-9d26-4813e8dc9745"
Operator feedback on a step feeds an assistant that knows the runbook and may diagnose the deployment (read-only access first), giving a feedback, retry, feedback, confirm loop. Needs decisions on access, identity and what is sent where.
2026-10-03: human gate routed to hub decision 1e517449-2114-4a8f-9c6b-fcdbe977378e (D9: runbook-tutorials execution-mode decisions (RBT-WP-0006 T01–T07)) by CUST-WP-0075; resolving that decision raises C-40 here.
Dependencies between runbooks
id: RBT-WP-0006-T04
status: wait
needs_human: true
decision_id: 1e517449-2114-4a8f-9c6b-fcdbe977378e
blocking_reason: "Needs a decision on how a pack in another repo is identified and trusted."
priority: low
state_hub_task_id: "acd969c3-3bd4-5953-9f9a-8ad8c5d8e380"
Packs declare requires_packs with a required result (exercised, or a named gate
passed), instead of deeper nesting (v0.2 caps nesting at three levels). Needs a decision
on how a pack in another repo is identified and trusted.
2026-10-03: human gate routed to hub decision 1e517449-2114-4a8f-9c6b-fcdbe977378e (D9: runbook-tutorials execution-mode decisions (RBT-WP-0006 T01–T07)) by CUST-WP-0075; resolving that decision raises C-40 here.
Terminal walk-through
id: RBT-WP-0006-T05
status: wait
needs_human: true
decision_id: 1e517449-2114-4a8f-9c6b-fcdbe977378e
blocking_reason: "Needs a decision whether hosts without a browser justify it; the CLI already covers v0.1 runs (moved from RBT-WP-0004-T03)."
priority: low
state_hub_task_id: "39750f3e-4e3c-56d5-b6ed-a9c6430ecadb"
rtut run <pack> walking a run in the terminal, for hosts without a browser. v0.2 runs can be driven from the engine and the UI, not yet from a terminal.
2026-10-03: human gate routed to hub decision 1e517449-2114-4a8f-9c6b-fcdbe977378e (D9: runbook-tutorials execution-mode decisions (RBT-WP-0006 T01–T07)) by CUST-WP-0075; resolving that decision raises C-40 here.
v0.3: close the console gaps
id: RBT-WP-0006-T06
status: wait
needs_human: true
decision_id: 1e517449-2114-4a8f-9c6b-fcdbe977378e
blocking_reason: "Needs a decision to start a v0.3 design: conditional applicability (G1), longer horizons (G2), gate recheck (G3), value predicates (G4), local-only attributes (G5), outputs feeding later members (G7). G6 is T01."
priority: medium
state_hub_task_id: "848f86a1-04f3-51cc-ae95-062f8694dd48"
The gaps in docs/console-mapping.md. Closing G1, G3 and G7 is also the precondition for extracting the console's engine parts and for modeling the
security bootstrap as a project (the cancelled RBT-WP-0005-T04).
2026-10-03: human gate routed to hub decision 1e517449-2114-4a8f-9c6b-fcdbe977378e (D9: runbook-tutorials execution-mode decisions (RBT-WP-0006 T01–T07)) by CUST-WP-0075; resolving that decision raises C-40 here.
Security and code review
id: RBT-WP-0006-T07
status: wait
needs_human: true
decision_id: 1e517449-2114-4a8f-9c6b-fcdbe977378e
blocking_reason: "Optional follow-up agreed with Bernd on 2026-09-29: the workplans close without it. Needs a decision when to run it."
priority: medium
state_hub_task_id: "4a89e7c1-b12e-5d50-9e23-fbd361c669da"
Run the security review and the code review over the engine and the UI: it serves files (artefacts, records), writes YAML (projects, attribute sets),
uses shell=True for operator-confirmed verify commands, and takes form input on localhost. Also first execution of .forgejo/workflows/tests.yaml on a real runner.
2026-10-03: status corrected by CUST-WP-0075 — active → blocked: all seven open tasks are wait with needs_human: true (each awaits a decision from Bernd), so no task is actionable.
2026-10-03: human gate routed to hub decision 1e517449-2114-4a8f-9c6b-fcdbe977378e (D9: runbook-tutorials execution-mode decisions (RBT-WP-0006 T01–T07)) by CUST-WP-0075; resolving that decision raises C-40 here.