Story
As a initiative owner,
I want to review the comparative pilot report and record a go/no-go decision, the selected server (if go), and the fleet-wide rollout recommendation,
so that activation beyond the pilot is a deliberate, documented human decision rather than an implicit drift from the pilot wiring.
Acceptance Criteria
- A decision record (in docs/lsp-pilot.md or a linked doc) states GO or NO-GO with the metric evidence from story 5's report cited.
- If GO: the selected LSP-MCP server is named with its pinned version, and the next-step scope is stated (e.g. extend to TypeScript in
markets, or to dev-lead code-writing) WITHOUT enabling it in this epic.
- If NO-GO: the pilot wiring is left inert (REVIEW_MCP_CONFIG unset for LSP) so default review behavior is unchanged, and the reasons are recorded for future re-evaluation.
- The decision explicitly confirms no fleet-wide or dev-lead LSP enablement happened as part of this epic (that is a separate, future initiative).
Tasks / Subtasks
Dev Notes
- This is a human decision (hands-off): activation beyond the pilot is never auto-released, consistent with how the epic itself is created inert.
- Grounding: the pilot deliberately stops at pr-review/Shell; dev-lead code-writing LSP and cross-language (markets TS) are out of scope here and must remain so unless a follow-up initiative is opened.
- No code change required beyond the decision record; if NO-GO, simply ensure the LSP MCP config is not referenced by REVIEW_MCP_CONFIG so the engine threads no LSP flags (behavior reverts byte-for-byte).
Project Structure Notes
Decision record only; no engine/workflow behavior change.
References
- docs/initiatives/mcp-powered-review.md
- .github/review-mcp.json
Likely target surface
docs/lsp-pilot.md (new file)
Story prepared by the BMAD Scrum Master (Bob) for epic #839. Status: ready-for-dev.
Story
As a initiative owner,
I want to review the comparative pilot report and record a go/no-go decision, the selected server (if go), and the fleet-wide rollout recommendation,
so that activation beyond the pilot is a deliberate, documented human decision rather than an implicit drift from the pilot wiring.
Acceptance Criteria
markets, or to dev-lead code-writing) WITHOUT enabling it in this epic.Tasks / Subtasks
Dev Notes
Project Structure Notes
Decision record only; no engine/workflow behavior change.
References
Likely target surface
docs/lsp-pilot.md (new file)Story prepared by the BMAD Scrum Master (Bob) for epic #839. Status: ready-for-dev.