Why: verify configured Agent CLIs one profile at a time before declaring them usable
Evidence: 53 focused tests, agent-exec validation, 397-test governance regression, and diff check
Impact: adds opt-in source-external probes, acceptance checks, and current task records
Refs: U5; docs/superpowers/plans/2026-10-15-project-agent-execution.md
Why: Hide provider-specific noninteractive, model, permission, workspace, prompt, and output differences behind one adapter seam.
Evidence: Cursor observed red then 22 green config/adapter tests; the host replays them and whitespace checks during integration.
Impact: Adds deterministic direct-argv adapters and probes without launching providers or enabling project routing.
Refs: docs/superpowers/plans/2026-10-15-project-agent-execution.md
Why: Establish one validated project-local routing document before any real Agent execution is enabled.
Evidence: Cursor observed red then 12 green configuration tests; the host replays those tests, the temporary validation target, and whitespace checks during integration.
Impact: Adds the U1 configuration contract and parser only; real routing, adapters, controller execution, and governance adoption remain disabled.
Refs: docs/superpowers/plans/2026-10-15-project-agent-execution.md
Why: Turn the approved multi-CLI routing design into independently testable Cursor implementation units without another manual plan gate.
Evidence: The plan self-review found R1-R10 coverage, stable interfaces, five review-focus tests, and no placeholders or whitespace errors.
Impact: Authorizes U1-U5 execution through fresh Cursor Agent runs while the host retains scope, verification, integration, and delivery control.
Refs: docs/superpowers/specs/2026-10-15-project-agent-execution-design.md
Why: Let each project route expensive implementation work across available Agent CLIs while the host model retains planning and delivery control.
Evidence: The specification passed placeholder, consistency, scope, path, and whitespace self-checks; implementation and runtime validation remain NOT_RUN.
Impact: Defines a review-only R1-R10 design for Markdown configuration, adapters, run-everything execution, receipts, recovery, and host fallback without enabling the new route.
Refs: .task/feature-goal-first-delivery.md
Why: Preserve the stage-six approval, resolved synchronization result, and exclusive final-window claim before fixing the tested candidate.
Evidence: make task-sync incorporated origin/main at cc40e10 with changed=0 after the conflict-resolution merge; queue claim is issue #518 comment 6478.
Impact: Advances this task to formal verification without changing implementation behavior or expanding its scope.
Refs: .task/feature-goal-first-delivery.md
Why: Synchronize the approved policy candidate with the active simplified merge-window rules now on origin/main.
Evidence: Resolved the two semantic conflicts by preserving upstream queue simplification and this task's goal-first/cursor-per-unit contract; focused task-card regression and cached diff checks passed.
Impact: Retains the active single-window delivery process while replacing mechanical stage approvals with necessary decisions and preserving Cursor-per-unit execution.
Refs: .task/feature-goal-first-delivery.md
Why: branches merged into main had left their task cards open; backfill status and archive to prevent bookkeeping debt.
Evidence: feature-merge-queue-simple-policy.md(PR#554) — merges verified via git log --merges plus Gitea API squash lookup against origin/main.
Impact: .task/*.md moved to .task/archive/ with status backfilled; no code or submodule change.
Refs: task.sh close (card-autoclose-r1)
Why: include the current integration baseline before verification.
Evidence: fetched origin/main=803a73e5d6621f7b5aac792f81f0fd449bd00be2.
Impact: integrate upstream changes without rewriting task history.
Refs: task-sync
Why: Preserve the one changed-base sync before freezing the final candidate.
Evidence: task-sync merged main b8edf13 without conflicts and reported tested checkpoint 4272d55.
Impact: Task record only; policy content and runtime behavior are unchanged.
Refs: Agent-System-Dev #518#554
Why: Keep the task authority, execution attempts, and stage-six candidate state recoverable.
Evidence: make evidence captured the controller status for P2 at canonical commit d79d6c17 and the run-wide governance verification receipt.
Impact: Records one failed no-change attempt, one successful Cursor attempt, host integration, and all remaining NOT_RUN stages without changing implementation behavior.
Refs: .task/feature-goal-first-delivery.md
Why: Ensure implementation authorship is delegated per unit while the primary model remains responsible for planning and delivery management.
Evidence: The focused task-card generation regression went red then green, and git diff --check passed.
Impact: Every repository-writing implementation unit gets a fresh Cursor CLI run; unavailable or failed Cursor work is recorded as a blocker instead of being silently implemented by the primary model.
Refs: .task/feature-goal-first-delivery.md
Why: include the current integration baseline before verification.
Evidence: fetched origin/main=b8edf13dc7b818d6e61b829f6e88202ec553f2fc.
Impact: integrate upstream changes without rewriting task history.
Refs: task-sync
Why: branches merged into main had left their task cards open; backfill status and archive to prevent bookkeeping debt.
Evidence: fix-8788-message-permission.md(PR#553) — merges verified via git log --merges plus Gitea API squash lookup against origin/main.
Impact: .task/*.md moved to .task/archive/ with status backfilled; no code or submodule change.
Refs: task.sh close (card-autoclose-r1)
Why: Keep authorized work focused on its stated outcome, prevent private scope expansion, and remove mechanical human stage gates.
Evidence: make test-governance passed 344 tests with 18 environment-dependent skips; receipt is recorded in the task card.
Impact: New tasks use goal-first execution while retaining real decision points, independent review, evidence, platform admission, and high-consequence dual keys.
Refs: .task/feature-goal-first-delivery.md
Why: make PR #553 the durable task record before final CI and merge.
Evidence: the Gitea API reports PR #553 open and mergeable at head 3138221.
Impact: task metadata only; the next push creates the final PR candidate.
Refs: .task/fix-8788-message-permission.md; PR #553
Why: preserve the exact pre-PR baseline and complete local regression output.
Evidence: task-sync fixed origin/main at 3326192; make test passed all required local checks.
Impact: task evidence only; no source, component pin, or runtime configuration change.
Refs: .task/fix-8788-message-permission.md
Why: Keep final sync and conflict handling with the merge agent and avoid treating pre-freeze SELF checks as final evidence.
Evidence: Review found one stale anchor, one ownership ambiguity, and one evidence-scope overclaim; all three are corrected.
Impact: Documentation only; no runtime or platform behavior changes.
Refs: Agent-System-Dev #518
Why: establish the authorized task workspace and its initial baseline.
Evidence: preserved precheck receipt in .task/feature-goal-first-delivery.md; R0=33261925a10c4eee40d1b6175b3895e78a47ef02.
Impact: task metadata only; the feature workspace starts at 33261925a1.
Refs: .task/feature-goal-first-delivery.md
Why: the user confirmed normal conversation behavior and authorized task closeout.
Evidence: user acceptance on 2026-10-15 supplements the preserved SELF runtime checks.
Impact: advances the task record to integration preparation; no production code change.
Refs: .task/fix-8788-message-permission.md
Why: Freeze the queue attempt and exact base before final validation.
Evidence: task-sync returned changed=0 on base 33261925; Issue #518 comment 6408 records the window.
Impact: Task record only; policy content is unchanged.
Refs: Agent-System-Dev #518
Why: bind the active runtime repair, rollback point, and review state to the synchronized task record.
Evidence: SELF runtime checks passed for message delivery, completed run, privacy, cleanup, credential revocation, and final database state.
Impact: task metadata only; user browser acceptance and task cleanup remain pending.
Refs: .task/fix-8788-message-permission.md
Why: Keep branches developing independently and serialize only final sync, validation, and agent merge without expanding into audits or new platform mechanisms.
Evidence: scope and link checks, git diff --check, and recursive worktree verification passed as SELF evidence in the task card.
Impact: Documentation only; existing task-sync, CI, branch protection, and agent merge behavior remain unchanged.
Refs: Agent-System-Dev #518
Why: include the current integration baseline before verification.
Evidence: fetched origin/main=33261925a10c4eee40d1b6175b3895e78a47ef02.
Impact: integrate upstream changes without rewriting task history.
Refs: task-sync
Why: preserve the authorized runtime role repair and its exact local evidence.
Evidence: recorded the pre-fix 404, role commands, completed test run, privacy check, backup, and revoked temporary credentials.
Impact: task metadata only; the live 8788 role configuration is documented without source changes.
Refs: .task/fix-8788-message-permission.md
Why: establish the authorized task workspace and its initial baseline.
Evidence: preserved precheck receipt in .task/feature-merge-queue-simple-policy.md; R0=33261925a10c4eee40d1b6175b3895e78a47ef02.
Impact: task metadata only; the feature workspace starts at 33261925a1.
Refs: .task/feature-merge-queue-simple-policy.md
Why: establish the authorized task workspace and its initial baseline.
Evidence: preserved precheck receipt in .task/fix-8788-message-permission.md; R0=99119a733830bc120cd6747b8b61736ea6da6078.
Impact: task metadata only; the feature workspace starts at 99119a7338.
Refs: .task/fix-8788-message-permission.md
Why: the close commit still said the live 8788 check had not run, which contradicts the user acceptance already recorded in the same card.
Evidence: user confirmation on 2026-10-15 of conversation order, pane deduplication, and folded process review; cutover notes already in the card. Local checks stay SELF.
Impact: task-card text only. No code, schema, or runtime change.
Refs: Dev PR #550, close PR #552
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: branches merged into main had left their task cards open; backfill status and archive to prevent bookkeeping debt.
Evidence: feature-work-companion-visibility.md(PR#550) — merges verified via git log --merges plus Gitea API squash lookup against origin/main.
Impact: .task/*.md moved to .task/archive/ with status backfilled; no code or submodule change.
Refs: task.sh close (card-autoclose-r1)
Why: include the current integration baseline before verification.
Evidence: fetched origin/main=bc76b68cd075379122430acb24fa080b0a38ff5e.
Impact: integrate upstream changes without rewriting task history.
Refs: task-sync
Why: Main advanced during UI delivery and task metadata must identify the new combined candidate.
Evidence: task-sync source-out receipt 6122ca6; Backend ead5ade includes both reviewed run history and concurrent vendor skills.
Impact: Fill the task PR link and mark the new Backend combination and final checks pending.
Refs: Agent-System-Dev #550; Agent-System-Backend #199 and #200
Why: Dev main advanced with a vendor skills task while this UI combination was being verified.
Evidence: Backend main ead5ade contains both Backend #199 and #200; the component merge affects recipes, not run-history handlers.
Impact: Preserve the new main task card and pin the Backend merge containing both changes; combination checks follow on this head.
Refs: Agent-System-Dev #550; Agent-System-Backend #199 and #200
Why: branches merged into main had left their task cards open; backfill status and archive to prevent bookkeeping debt.
Evidence: feature-vendor-popular-skills.md(PR#549) — merges verified via git log --merges plus Gitea API squash lookup against origin/main.
Impact: .task/*.md moved to .task/archive/ with status backfilled; no code or submodule change.
Refs: task.sh close (card-autoclose-r1)
Why: The fast and slow channel UI duplicated work cards and hid intermediate output after the final reply.
Evidence: SELF isolated-network make test and make build; 15 browser scenarios; independent component review; component PRs #55, #199, #85.
Impact: Pin merged Backend, Frontend and UI revisions and update DESIGN.md for one detail location and durable folded history.
Refs: .task/feature-work-companion-visibility.md; Agent-System-UI #55; Agent-System-Backend #199; Agent-System-Frontend #85
weichen
