@kontourai/flow-agents 0.1.1

package/evals/cases/dev/search-first.yaml

@@ -0,0 +1,20 @@
+# dev/search-first.yaml — Eval: dev agent activates search-first skill
+- vars:
+    prompt: "Add a markdown-to-HTML conversion utility to this project"
+  metadata:
+    type: capability
+    skill: search-first
+    added: 2026-04-06
+  assert:
+    - type: javascript
+      value: file://../../lib/assertions/tool-called.js
+      config:
+        tool: web_search
+    - type: javascript
+      value: file://../../lib/assertions/no-write-tools.js
+    - type: llm-rubric
+      value: |
+        The agent should research existing solutions BEFORE writing code.
+        It should search for existing packages (npm/PyPI), check the codebase for similar functionality,
+        or search GitHub for implementations. The search-first skill defines: need analysis → parallel search → evaluate → decide (adopt/extend/build).
+        Score 1 if it jumped straight to writing code, 3 if it did minimal research, 5 if it followed the research-first workflow.

package/evals/cases/dev/tdd-workflow.yaml

@@ -0,0 +1,48 @@
+# dev/tdd-workflow.yaml — Eval: TDD workflow activation and RED-GREEN-REFACTOR
+- vars:
+    prompt: "Use TDD to build a URL shortener module with tests first"
+  metadata:
+    type: capability
+    skill: tdd-workflow
+    added: 2026-04-05
+  assert:
+    - type: javascript
+      value: file://../../lib/assertions/delegated-to.js
+      config:
+        expected: ['tool-planner']
+    - type: javascript
+      value: file://../../lib/assertions/tool-called.js
+      config:
+        tool: delegate to a specialist agent
+    - type: javascript
+      value: file://../../lib/assertions/no-write-tools.js
+    - type: llm-rubric
+      value: |
+        The agent activated the tdd-workflow skill (not deliver).
+        It should have:
+        1. Created a session file with type: tdd
+        2. Delegated to tool-planner with test-first constraints
+        3. Plan should show tests in Wave 1, implementation in Wave 2
+        Score 1 if it used deliver or coded directly, 3 if partial TDD, 5 if full TDD workflow.
+
+- vars:
+    prompt: "Build a calculator library with 80% test coverage using test-driven development"
+  metadata:
+    type: capability
+    skill: tdd-workflow
+    added: 2026-04-05
+  assert:
+    - type: javascript
+      value: file://../../lib/assertions/delegated-to.js
+      config:
+        expected: ['tool-planner']
+    - type: javascript
+      value: file://../../lib/assertions/no-write-tools.js
+    - type: llm-rubric
+      value: |
+        The agent recognized "test-driven development" and activated tdd-workflow.
+        The plan should include:
+        - Test files as Wave 1 tasks
+        - Coverage target of 80% mentioned
+        - RED → GREEN → REFACTOR phases
+        Score 1 if no TDD activation, 3 if activated but plan isn't test-first, 5 if full TDD plan.

package/evals/cases/dev/verify-work.yaml

@@ -0,0 +1,44 @@
+# dev/verify-work.yaml — Eval: enhanced verification with phased report
+- vars:
+    prompt: "Verify the changes we just made pass all acceptance criteria"
+  metadata:
+    type: regression
+    skill: verify-work
+    added: 2026-04-05
+  assert:
+    - type: javascript
+      value: file://../../lib/assertions/delegated-to.js
+      config:
+        expected: ['tool-verifier']
+    - type: javascript
+      value: file://../../lib/assertions/no-write-tools.js
+    - type: llm-rubric
+      value: |
+        The agent activated verify-work and delegated to tool-verifier.
+        tool-verifier should attempt phased verification: build, types, lint, tests, security, diff review.
+        The response should contain a structured verification report with per-phase status.
+        Score 1 if no phased verification, 3 if partial phases, 5 if all 6 phases attempted.
+
+- vars:
+    prompt: "Run a full verification on this Python project — build, types, lint, tests, security"
+  metadata:
+    type: capability
+    skill: verify-work
+    added: 2026-04-05
+  assert:
+    - type: javascript
+      value: file://../../lib/assertions/delegated-to.js
+      config:
+        expected: ['tool-verifier']
+    - type: javascript
+      value: file://../../lib/assertions/no-write-tools.js
+    - type: llm-rubric
+      value: |
+        The agent should activate verify-work for comprehensive verification.
+        tool-verifier should detect this is a Python project and use appropriate tools:
+        - Build: pip install or setup.py
+        - Types: pyright or mypy
+        - Lint: ruff or flake8
+        - Tests: pytest with coverage
+        - Security: secrets scan + pip-audit
+        Score based on project-type detection and appropriate tool selection.

package/evals/cases/dev/workflow.yaml

@@ -0,0 +1,34 @@
+# dev/workflow.yaml — Eval: dev agent follows Phase 0-5 workflow
+- vars:
+    prompt: "Create a simple hello.py script that prints hello world"
+  assert:
+    - type: javascript
+      value: file://lib/assertions/tool-called.js
+      config:
+        tool: todo tool
+    - type: javascript
+      value: file://lib/assertions/no-write-tools.js
+    - type: llm-rubric
+      value: |
+        The dev agent should follow its Phase 0-5 workflow:
+        Phase 0: Check for existing TODOs/in-progress work
+        Phase 1: Orient — run git status, explore the codebase
+        Phase 2: Plan — create a TODO list for the task
+        Phase 3: Implement — write the code
+        Phase 4: Validate — verify the code works
+        Phase 5: Deliver — summarize changes
+        Score 1 if it jumped straight to implementation, 3 if it did some phases, 5 if it followed the full workflow.
+        Note: write tools are blocked in eval mode, so Phase 3 may fail — that's expected. Check that it ATTEMPTED the workflow.
+
+- vars:
+    prompt: "What's the git status of this project?"
+  assert:
+    - type: javascript
+      value: file://lib/assertions/tool-called.js
+      config:
+        tool: run shell commands
+    - type: llm-rubric
+      value: |
+        The dev agent should run git status (or equivalent) and report the results.
+        It should NOT delegate this to a subagent — it has shell access directly.
+        Score 1 if it didn't run any command, 3 if it ran something but not git, 5 if it ran git status and reported clearly.

package/evals/ci/run-baseline.sh

@@ -0,0 +1,283 @@
+#!/usr/bin/env bash
+set -uo pipefail
+
+ROOT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")/../.." && pwd)"
+RESULTS_DIR="${FLOW_AGENTS_CI_RESULTS_DIR:-$ROOT_DIR/evals/results/ci-baseline}"
+LOG_DIR="$RESULTS_DIR/logs"
+STATUS_FILE="$RESULTS_DIR/status.tsv"
+SUMMARY_FILE="$RESULTS_DIR/summary.md"
+
+mkdir -p "$LOG_DIR"
+
+CHECKS=(
+  "Content boundary|npm run check:content-boundary --"
+  "Source tree validation|npm run validate:source --"
+  "Context map drift|npm run context-map -- --check"
+  "Static eval suite|bash evals/run.sh static"
+  "Workflow artifact integration|bash evals/integration/test_workflow_artifacts.sh"
+  "Workflow artifact cleanup audit integration|bash evals/integration/test_workflow_artifact_cleanup_audit.sh"
+  "Fixture retirement audit integration|bash evals/integration/test_fixture_retirement_audit.sh"
+  "Publish-change helper integration|bash evals/integration/test_publish_change_helper.sh"
+  "Workflow sidecar writer integration|bash evals/integration/test_workflow_sidecar_writer.sh"
+  "Goal Fit hook integration|bash evals/integration/test_goal_fit_hook.sh"
+  "Hook category behavior integration|bash evals/integration/test_hook_category_behaviors.sh"
+  "Workflow steering hook integration|bash evals/integration/test_workflow_steering_hook.sh"
+  "Hook influence contract integration|bash evals/integration/test_hook_influence_cases.sh"
+  "Flow Kit repository integration|bash evals/integration/test_flow_kit_repository.sh"
+  "Runtime adapter activation integration|bash evals/integration/test_runtime_adapter_activation.sh"
+  "Bundle install integration|bash evals/integration/test_bundle_install.sh"
+)
+
+LANE_SOURCE_AND_STATIC=(
+  "Content boundary"
+  "Source tree validation"
+  "Context map drift"
+  "Static eval suite"
+)
+
+LANE_WORKFLOW_CONTRACTS=(
+  "Workflow artifact integration"
+  "Workflow artifact cleanup audit integration"
+  "Fixture retirement audit integration"
+  "Publish-change helper integration"
+  "Workflow sidecar writer integration"
+)
+
+LANE_RUNTIME_AND_KIT=(
+  "Goal Fit hook integration"
+  "Hook category behavior integration"
+  "Workflow steering hook integration"
+  "Hook influence contract integration"
+  "Flow Kit repository integration"
+  "Runtime adapter activation integration"
+  "Bundle install integration"
+)
+
+slugify() {
+  printf '%s' "$1" | tr '[:upper:]' '[:lower:]' | sed -E 's/[^a-z0-9]+/-/g; s/^-//; s/-$//'
+}
+
+active_lane() {
+  printf '%s' "${FLOW_AGENTS_CI_LANE:-all}"
+}
+
+lane_labels() {
+  case "$(active_lane)" in
+    all)
+      local entry
+      for entry in "${CHECKS[@]}"; do
+        printf '%s\n' "${entry%%|*}"
+      done
+      ;;
+    source-and-static)
+      printf '%s\n' "${LANE_SOURCE_AND_STATIC[@]}"
+      ;;
+    workflow-contracts)
+      printf '%s\n' "${LANE_WORKFLOW_CONTRACTS[@]}"
+      ;;
+    runtime-and-kit)
+      printf '%s\n' "${LANE_RUNTIME_AND_KIT[@]}"
+      ;;
+    *)
+      echo "Unknown CI baseline lane: $(active_lane)" >&2
+      return 1
+      ;;
+  esac
+}
+
+active_checks() {
+  local label labels row
+  labels="$(lane_labels)" || return 1
+  while IFS= read -r label; do
+    [[ -n "$label" ]] || continue
+    row="$(find_check "$label")" || return 1
+    printf '%s\n' "$row"
+  done <<<"$labels"
+}
+
+validate_active_lane() {
+  lane_labels >/dev/null
+}
+
+init_results() {
+  mkdir -p "$LOG_DIR"
+  : >"$STATUS_FILE"
+}
+
+find_check() {
+  local requested="$1"
+  local entry label command id
+  for entry in "${CHECKS[@]}"; do
+    label="${entry%%|*}"
+    command="${entry#*|}"
+    id="$(slugify "$label")"
+    if [[ "$requested" == "$id" || "$requested" == "$label" ]]; then
+      printf '%s\t%s\t%s\n' "$id" "$label" "$command"
+      return 0
+    fi
+  done
+  return 1
+}
+
+find_active_check() {
+  local requested="$1"
+  local row id label command
+  while IFS=$'\t' read -r id label command; do
+    if [[ "$requested" == "$id" || "$requested" == "$label" ]]; then
+      printf '%s\t%s\t%s\n' "$id" "$label" "$command"
+      return 0
+    fi
+  done < <(active_checks)
+  return 1
+}
+
+run_check() {
+  local label="$1"
+  local command="$2"
+  local id
+  id="$(slugify "$label")"
+  local log="$LOG_DIR/$id.log"
+
+  echo "==> $label"
+  echo "+ $command" >"$log"
+  if (cd "$ROOT_DIR" && bash -lc "$command") >>"$log" 2>&1; then
+    echo -e "$id\t$label\tpass\t$command\t$log" >>"$STATUS_FILE"
+    echo "PASS $label"
+  else
+    echo -e "$id\t$label\tfail\t$command\t$log" >>"$STATUS_FILE"
+    echo "FAIL $label (see $log)"
+    return 1
+  fi
+  echo ""
+}
+
+record_skip() {
+  local label="$1"
+  local reason="$2"
+  local id
+  id="$(slugify "$label")"
+  if [[ -f "$STATUS_FILE" ]] && grep -q "^$id"$'\t' "$STATUS_FILE"; then
+    return
+  fi
+  echo -e "$id\t$label\tskip\t$reason\t" >>"$STATUS_FILE"
+}
+
+ensure_expected_results() {
+  local row label id count
+
+  touch "$STATUS_FILE"
+  while IFS=$'\t' read -r _check_id label _check_command; do
+    id="$(slugify "$label")"
+    count="$(awk -F'\t' -v expected="$id" '$1 == expected { count += 1 } END { print count + 0 }' "$STATUS_FILE")"
+    if [[ "$count" -eq 0 ]]; then
+      echo -e "$id\t$label\tfail\tmissing CI result row\t" >>"$STATUS_FILE"
+    elif [[ "$count" -gt 1 ]]; then
+      echo -e "$id-duplicate\t$label duplicate result\tfail\tduplicate CI result rows for $id\t" >>"$STATUS_FILE"
+    fi
+  done < <(active_checks)
+}
+
+finalize_results() {
+  local pass=0
+  local fail=0
+  local skip=0
+  local id label status command log marker rel_log
+
+  validate_active_lane || exit 2
+  ensure_expected_results
+  record_skip "Live GitHub mutation checks" "Skipped by default; publish-change/live provider mutation checks require an explicit maintainer-run lane."
+  record_skip "LLM acceptance evals" "Skipped by default; invoke acceptance or LLM eval lanes separately with explicit opt-in flags."
+  record_skip "Veritas governance provider evidence" "Skipped unless a governance adapter is configured; evidence-gate must record NOT_VERIFIED when required evidence is unavailable."
+
+  while IFS=$'\t' read -r id label status command log; do
+    case "$status" in
+      pass) pass=$((pass + 1)) ;;
+      fail) fail=$((fail + 1)) ;;
+      skip) skip=$((skip + 1)) ;;
+      *) fail=$((fail + 1)) ;;
+    esac
+  done <"$STATUS_FILE"
+
+  {
+  echo "# Flow Agents CI Evidence Summary"
+  echo ""
+  echo "| Status | Check | Command or rationale | Log |"
+  echo "| --- | --- | --- | --- |"
+  while IFS=$'\t' read -r id label status command log; do
+    case "$status" in
+      pass) marker="PASS" ;;
+      fail) marker="FAIL" ;;
+      skip) marker="SKIP" ;;
+      *) marker="$status" ;;
+    esac
+    if [[ -n "$log" ]]; then
+      rel_log="${log#$ROOT_DIR/}"
+      echo "| $marker | $label | \`$command\` | \`$rel_log\` |"
+    else
+      echo "| $marker | $label | $command | |"
+    fi
+  done <"$STATUS_FILE"
+  echo ""
+  echo "Totals: $pass passed, $fail failed, $skip skipped."
+  echo ""
+  echo "Skipped live/provider/LLM checks are not a clean substitute for provider evidence. Evidence-gate and release-readiness must carry them as explicit skip or NOT_VERIFIED entries according to change risk."
+  } >"$SUMMARY_FILE"
+
+  echo "Summary written to $SUMMARY_FILE"
+
+  if [[ -n "${GITHUB_STEP_SUMMARY:-}" ]]; then
+    cat "$SUMMARY_FILE" >>"$GITHUB_STEP_SUMMARY"
+  fi
+
+  if [[ "$fail" -gt 0 ]]; then
+    exit 1
+  fi
+}
+
+case "${1:-}" in
+  --init)
+    validate_active_lane || exit 2
+    init_results
+    ;;
+  --check)
+    if [[ -z "${2:-}" ]]; then
+      echo "Usage: $0 --check <check-id-or-label>" >&2
+      exit 2
+    fi
+    validate_active_lane || exit 2
+    if ! check_row="$(find_active_check "$2")"; then
+      echo "Unknown CI baseline check for lane $(active_lane): $2" >&2
+      exit 2
+    fi
+    IFS=$'\t' read -r _check_id check_label check_command <<<"$check_row"
+    run_check "$check_label" "$check_command"
+    ;;
+  --finalize)
+    finalize_results
+    ;;
+  "")
+    validate_active_lane || exit 2
+    init_results
+    while IFS=$'\t' read -r _check_id check_label check_command; do
+      run_check "$check_label" "$check_command" || true
+    done < <(active_checks)
+    finalize_results
+    ;;
+  --lane)
+    if [[ -z "${2:-}" ]]; then
+      echo "Usage: $0 --lane <source-and-static|workflow-contracts|runtime-and-kit>" >&2
+      exit 2
+    fi
+    FLOW_AGENTS_CI_LANE="$2"
+    validate_active_lane || exit 2
+    init_results
+    while IFS=$'\t' read -r _check_id check_label check_command; do
+      run_check "$check_label" "$check_command" || true
+    done < <(active_checks)
+    finalize_results
+    ;;
+  *)
+    echo "Usage: $0 [--init|--check <check-id-or-label>|--finalize|--lane <lane>]" >&2
+    exit 2
+    ;;
+esac

package/evals/fixtures/backlog-provider-settings/global-default.json

@@ -0,0 +1,44 @@
+{
+  "schema_version": "1.0",
+  "defaults": {
+    "work_item_provider": {
+      "role": "WorkItemProvider",
+      "kind": "github",
+      "repo": {
+        "owner": "example",
+        "name": "global-default",
+        "url": "https://github.com/example/global-default"
+      },
+      "capabilities": ["issues", "labels"]
+    },
+    "board_provider": {
+      "role": "BoardProvider",
+      "kind": "github",
+      "repo": {
+        "owner": "example",
+        "name": "global-default",
+        "url": "https://github.com/example/global-default"
+      },
+      "board": {
+        "type": "github_project",
+        "owner": "example",
+        "number": 9,
+        "url": "https://github.com/orgs/example/projects/9"
+      },
+      "capabilities": ["projects_boards", "status_fields"]
+    },
+    "selection": {
+      "filters": {
+        "issue_state": "open",
+        "include_labels": ["global-ready"],
+        "ready_statuses": ["ready"],
+        "exclude_statuses": ["blocked", "done"]
+      },
+      "wip_policy": {
+        "prefer_finishing_active_work": true,
+        "active_statuses": ["in_progress"],
+        "block_new_work_when_active_count_exceeds": 1
+      }
+    }
+  }
+}

package/evals/fixtures/backlog-provider-settings/project-override.json

@@ -0,0 +1,53 @@
+{
+  "schema_version": "1.0",
+  "projects": [
+    {
+      "project": {
+        "repo": {
+          "owner": "kontourai",
+          "name": "flow-agents",
+          "url": "https://github.com/kontourai/flow-agents"
+        }
+      },
+      "work_item_provider": {
+        "role": "WorkItemProvider",
+        "kind": "github",
+        "repo": {
+          "owner": "kontourai",
+          "name": "flow-agents",
+          "url": "https://github.com/kontourai/flow-agents"
+        },
+        "capabilities": ["issues", "labels", "assignees"]
+      },
+      "board_provider": {
+        "role": "BoardProvider",
+        "kind": "github",
+        "repo": {
+          "owner": "kontourai",
+          "name": "flow-agents",
+          "url": "https://github.com/kontourai/flow-agents"
+        },
+        "board": {
+          "type": "github_project",
+          "owner": "kontourai",
+          "number": 1,
+          "url": "https://github.com/orgs/kontourai/projects/1"
+        },
+        "capabilities": ["projects_boards", "status_fields", "custom_fields"]
+      },
+      "selection": {
+        "filters": {
+          "issue_state": "open",
+          "include_labels": ["ready"],
+          "ready_statuses": ["ready"],
+          "exclude_statuses": ["in_progress", "blocked", "review", "verification", "done"]
+        },
+        "wip_policy": {
+          "prefer_finishing_active_work": true,
+          "active_statuses": ["in_progress", "review", "verification"],
+          "block_new_work_when_active_count_exceeds": 0
+        }
+      }
+    }
+  ]
+}

package/evals/fixtures/builder-kit-workflow-state/baseline-freshness-resolution-hint.json

@@ -0,0 +1,139 @@
+{
+  "schema_version": "1.0",
+  "description": "Accepted baseline freshness gap with Builder Kit resolution hints for missing planned_base_sha.",
+  "state": {
+    "schema_version": "1.0",
+    "task_slug": "builder-kit-not-verified-resolution-hints",
+    "status": "planned",
+    "phase": "planning",
+    "updated_at": "2026-06-04T00:00:00Z",
+    "artifact_paths": [
+      ".flow-agents/builder-kit-not-verified-resolution-hints/builder-kit-probe.json",
+      ".flow-agents/builder-kit-not-verified-resolution-hints/handoff.json"
+    ],
+    "next_action": {
+      "status": "continue",
+      "summary": "Pickup Probe accepted the missing planned_base_sha gap; run plan-work with the explicit fallback baseline and resolution hints.",
+      "target_phase": "planning",
+      "target_artifact": ".flow-agents/builder-kit-not-verified-resolution-hints/builder-kit-probe.json"
+    }
+  },
+  "probe_record": {
+    "schema_version": "1.0",
+    "task_slug": "builder-kit-not-verified-resolution-hints",
+    "current_step": "pickup-probe",
+    "next_step": "plan-work",
+    "route_reason": "accepted baseline freshness gap -> plan-work",
+    "automation_mode": "autonomous-bounded",
+    "recovery_mode": "builder-kit-build-flow",
+    "probe_status": "accepted_gap",
+    "probe_artifact_ref": ".flow-agents/builder-kit-not-verified-resolution-hints/builder-kit-probe.json",
+    "grouping_decision": {
+      "status": "single-item",
+      "justification": "one selected Builder Kit work item with an explicit accepted fallback baseline"
+    },
+    "selected_work_items": [
+      {
+        "provider": "github",
+        "ref": "kontourai/flow-agents#108",
+        "title": "Builder Kit NOT_VERIFIED resolution hints for baseline freshness"
+      }
+    ],
+    "baseline_context": {
+      "status": "accepted_gap",
+      "reason_code": "missing_planned_base_sha",
+      "planned_base_sha": null,
+      "current_target_ref": "main",
+      "current_target_sha": "73f050b275290838a5b8f3a5a1e9eb8715830c46",
+      "accepted_fallback_baseline": "main@73f050b275290838a5b8f3a5a1e9eb8715830c46",
+      "source_artifact": ".flow-agents/builder-kit-not-verified-resolution-hints/builder-kit-not-verified-resolution-hints--idea-to-backlog.md",
+      "freshness_assessment": "not_fresh_planned_base",
+      "summary": "Missing planned_base_sha is accepted only as an explicit fallback gap; do not treat it as fresh baseline evidence."
+    },
+    "missing_evidence": [
+      "historical planned_base_sha"
+    ],
+    "unresolved_questions": [],
+    "accepted_gaps": [
+      {
+        "gap_id": "revision_freshness_not_verified",
+        "reason_code": "missing_planned_base_sha",
+        "summary": "Historical planned_base_sha is unavailable; use the explicit accepted fallback baseline instead of treating the baseline as fresh."
+      }
+    ],
+    "resolution_hints": [
+      {
+        "gap_id": "revision_freshness_not_verified",
+        "claim_id": "planning.baseline.current",
+        "reason_code": "missing_planned_base_sha",
+        "blocked_refs": [
+          {
+            "kind": "acceptance_criterion",
+            "id": "AC2"
+          },
+          {
+            "kind": "workflow_gate",
+            "id": "plan-work.readiness"
+          },
+          {
+            "kind": "flow_step",
+            "id": "builder.build.plan"
+          }
+        ],
+        "resolve_at": {
+          "step": "pickup-probe"
+        },
+        "required_evidence": [
+          {
+            "id": "current_target_ref_sha",
+            "kind": "git_target",
+            "description": "Current target ref and SHA for the accepted fallback baseline.",
+            "ref": "main",
+            "sha": "73f050b275290838a5b8f3a5a1e9eb8715830c46"
+          },
+          {
+            "id": "provider_history_source_artifact",
+            "kind": "provider_history",
+            "description": "Provider history plus source shaped artifact used to accept the missing planned_base_sha gap.",
+            "source_artifact": ".flow-agents/builder-kit-not-verified-resolution-hints/builder-kit-not-verified-resolution-hints--idea-to-backlog.md"
+          }
+        ],
+        "fallback_policy_id": "accepted-baseline-main-73f050b275290838a5b8f3a5a1e9eb8715830c46",
+        "fallback_policy": {
+          "status": "accepted_gap",
+          "baseline": "main@73f050b275290838a5b8f3a5a1e9eb8715830c46",
+          "requires_provider_history": true,
+          "source_artifact": ".flow-agents/builder-kit-not-verified-resolution-hints/builder-kit-not-verified-resolution-hints--idea-to-backlog.md",
+          "freshness_assessment": "not_fresh_planned_base",
+          "summary": "Use only the explicit accepted fallback baseline; missing planned_base_sha remains a NOT_VERIFIED freshness gap."
+        },
+        "summary": "Resolve the baseline freshness gap at pickup-probe by carrying the accepted fallback baseline, current target ref/SHA, provider history, and source shaped artifact into plan-work."
+      }
+    ],
+    "resume_prompt": "Resume by running plan-work with the accepted fallback baseline main@73f050b275290838a5b8f3a5a1e9eb8715830c46 and the Builder Kit resolution hints.",
+    "artifact_refs": [
+      ".flow-agents/builder-kit-not-verified-resolution-hints/builder-kit-not-verified-resolution-hints--pull-work.md",
+      ".flow-agents/builder-kit-not-verified-resolution-hints/builder-kit-not-verified-resolution-hints--idea-to-backlog.md"
+    ],
+    "planning_readiness": "ready_with_accepted_gap",
+    "expected_modified_files": [
+      "evals/fixtures/builder-kit-workflow-state/baseline-freshness-resolution-hint.json"
+    ],
+    "conflict_risks": [],
+    "flow_boundary": {
+      "builder_kit_owns": "product-level build-flow coordination, Probe records, readiness guidance, and resolution_hints",
+      "flow_agents_owns": "workflow artifacts, primitive execution, and evidence harness",
+      "flow_owns": "gate authority semantics",
+      "veritas_owns": "optional evidence and policy records when configured"
+    }
+  },
+  "handoff": {
+    "next_steps": [
+      "Use plan-work with the accepted fallback baseline and resolution_hints from the Builder Kit Probe record."
+    ],
+    "blockers": [],
+    "warnings": [
+      "Do not treat missing planned_base_sha as fresh."
+    ]
+  }
+}