@chenmk/superflow 0.4.0 → 0.5.1

package/assets/skills-en/superflow-pipeline/references/managed-work.md

@@ -1,34 +1,184 @@
 # Managed Work Rules
 
+## Design-constitution gate
+
+Before changing Superflow managed work itself, read
+`docs/managed-work-design-principles.en.md` and
+`docs/managed-agent-protocol.en.md` completely. Before delivery, walk through
+the constitution's mandatory change-review checklist and state whether the
+change alters any principle.
+
+Runner and scripts decide deterministic facts. Agents decide correctness,
+reasonableness, risk, and evidence sufficiency. Scripts must not replace
+semantic review with regexes, and Agents must not take ownership of hashes,
+schemas, counters, or state transitions. A new long-term rule first updates the
+Chinese and English constitutions, protocol, and tests.
+
+Managed orchestration owns only dual-Agent dispatch, frozen context, state,
+waiting, evidence, and review loops. API, database, concurrency, transaction,
+and test design belongs to `api.md`, `design.md`, `tests.md`, and source-level
+technical design. Executor prompts inherit rather than redesign. Audit every
+managed-prompt change against this boundary.
+
 Managed work is built into `superflow-pipeline`. The user only says to use
 Superflow and provides an implementation prompt, change directory, or direct
 task. Direct tasks receive a minimal contract. SDD work freezes the generated
 implementation prompt and never substitutes `tasks.md` for the execution prompt.
 
-Always dispatch through:
+Before start, the Host routes by input maturity: docs-only work stops at Coding
+Ready without a managed Task; a user-approved change/prompt is frozen directly;
+a bounded verbal task uses minimal or standard. If a verbal request has a
+direction-changing API, data, concurrency, cross-repository, or owner choice,
+return to clarification/documents instead of letting Runner regexes or Executor
+improvisation replace Host judgment.
+
+When MCP is installed, the current host must prefer the `superflow_managed_*`
+tools for submission, waiting, user guidance, and review submission. MCP is a
+channel to the local state machine rather than a third model. Direct host review
+is the only supported mode, preventing a nested supervisor CLI from duplicating
+context, tokens, and review time.
+
+When MCP is unavailable, dispatch through:
 
 ```bash
 superflow pipeline "<implementation-prompt|change-dir|task>" --managed --project "<root>" \
-  --supervisor <codex|claude> --executor <claude|codex> --language <en|zh>
+  --supervisor current --executor peer --language <en|zh>
 ```
 
-The command always starts the independent background service and follows the
-local event journal to a terminal state. Losing the follower does not stop the
-task. After return, the current agent reads the report and summarizes it in the
-original conversation.
+`current/peer` resolves the host and its counterpart automatically. If host
+detection is ambiguous, pass explicit `codex/claude` roles instead of guessing.
+Legacy contracts with missing or retired supervision settings are migrated to
+direct host review on resume. No entry point may spawn a background supervisor
+CLI.
+Keep a single foreground `pipeline`/`resume` waiter so the CLI reads local state;
+never spend host model turns periodically invoking `superflow status`.
+In MCP mode,
+`superflow_managed_wait(taskId, afterSequence, timeoutSeconds=240)` defaults to
+a Host-compatible transport window below common 300-second tool limits. Hosts
+that support longer calls may explicitly request up to 43200 seconds. An active
+request exposes stages and healthy checkpoints through standard MCP progress
+notifications without waking the Host model. One healthy or idle checkpoint does
+not complete wait; two consecutive checkpoints without an effective milestone
+return attention. Ordinary window expiry immediately continues wait from the
+compact snapshot's `latestSequence`, without status or Executor restart. Call
+`superflow_managed_status` for full evidence only when the state requires Host
+attention.
+
+After executor delivery, the task enters `external_supervisor_review_required`.
+The current host reads `host-review-N.md`, writes the review JSON, and resumes:
+
+```bash
+superflow pipeline --resume-task <task-id> --submit-host-review <review.json>
+```
 
 Hard gates:
 
-- Claim the 5/7/12 budgets before an Agent call.
+- Claim the 5/7/12 budgets before an Agent call. The default Executor circuit
+  also stops at two million cumulative token units. Provider-dollar accounting
+  is not comparable, so the default has no USD circuit and Claude receives no
+  `--max-budget-usd`.
 - Freeze the selected language into the contract; prompts, journals, reports, notifications, errors, and resumed rounds must use that same language.
 - Revalidate the frozen contract hash, non-overridable permissions, and 5/7/12 hard limits on every start and resume. Fail closed if persisted contract data changed.
+- Runner uses filesystem snapshots for non-Git workspace changes. Executors must
+  never run `git init` for diff, preflight, or delivery, and must not create or
+  replace `.git`; repository-identity changes during a task still fail closed.
+- The context manifest marks frozen requirements, designs, prompts, handoffs,
+  rules, and task facts as `immutable`, and task/test-report progress evidence
+  as `retain`. Immutable entries are read-only; retained entries may be updated
+  but never deleted. Native hooks block before mutation and every Host is still
+  verified by preflight/final gates.
+- Cleanup separates runtime, workspace-temporary, delivery-artifacts, and
+  protected-contracts. Executor cleans only the first two and retains source,
+  SQL, tests, reports, and frozen contracts for Host review. Never use pkill,
+  killall, a bare kill, port/name-only cleanup, or rm -rf without owner-helper verification.
 - Resolve change directories through `.sdd/state.yaml` `implementation_prompt`; treat `tasks.md` as a checklist only. Copy the prompt into a managed snapshot and freeze its SHA-256 for both agents.
-- Persist and resume exact supervisor and executor session IDs.
+- The first executor invocation automatically uses the frozen prompt as its task entry.
+- For real-runtime work, the first prompt requires a repository-owned,
+  one-command, repeatable, failure-safe, portable acceptance entry covering
+  preflight, setup, build, startup, real calls, assertions, and cleanup,
+  including the success path and one representative setup/verification failure
+  cleanup with final zero residue. The Host reviews all properties together.
+- Call the MCP runtime handshake before start. An installed-runtime change makes
+  the stale MCP fail closed before persistence and requires a Host restart.
+  Persist verbose logs and print bounded summaries and failure tails only.
+- Verification has three levels: repairs use change-scoped checks; engineering/SDD
+  defaults to task-level real acceptance; only an explicit frozen prompt, an owner
+  helper/managed-orchestration change, or a new custom cleanup primitive uses
+  framework certification. The frozen contract may raise but never silently lower
+  the level, and historical evidence reruns only when the diff/finding invalidates it.
+- A normal real-runtime Host review checks task PID/start-fingerprint binding,
+  deletion gates, one representative setup/assertion failure cleanup, and final
+  zero residue. Distinct business deletion paths and affected/cross-service
+  repositories still receive their own proof. It must not rerun shared-helper PID-reuse, non-owner, forged-state,
+  and every-signal disaster suites for each business task. Reuse the helper directly
+  through a configurable path instead of copying a cleanup framework. Record Host
+  usage or an explicit unavailable reason.
+- Pass neither `--max-turns` nor `--max-budget-usd` to a Claude Executor. Do not
+  split normal work by a fixed tool-step count or an inaccurate provider-dollar
+  estimate. The defaults stop after 60 minutes with no output or a two-hour
+  single-invocation deadline; repeated provider failures, token/invocation
+  budgets, safety boundaries, and user pause still stop execution.
+- Claude Executor defaults to the official
+  `CLAUDE_CODE_AUTO_COMPACT_WINDOW=200000` and
+  `CLAUDE_AUTOCOMPACT_PCT_OVERRIDE=70` settings for earlier long-session
+  compaction; explicit user values win. It reloads the frozen prompt and handoff
+  after compaction, and final `result.usage` is authoritative for the invocation.
+- Network errors, closed connections, timeouts, and provider overloads share one
+  consecutive-failure counter and must open the circuit on the third failure;
+  only a complete Agent response resets it. Transient 429/529/overloaded calls
+  and connection refused/reset/close or socket failures confirmed before an Agent
+  response may be credited; permission, token-plan, quota, and billing failures
+  are not, and none may bypass the three-failure circuit.
+- Keep the first invocation in one continuous session. Create a new session for
+  review repair or abnormal recovery and retain old IDs for audit; never guess
+  with `--last` or `--continue`. Generate `executor-handoff-N.md` for later
+  invocations with the prompt hash, task progress, local pending work, external
+  prerequisites, all findings, diff stat, and recent report. A new session must
+  read it and verify against the current workspace.
+- Write safely discovered Java, Maven, Node, and npm executable paths into the
+  first executor policy so the Agent does not waste tool calls rediscovering the
+  basic toolchain.
+- Handoff includes exact applicable rule files and a deterministic delivery
+  preflight. Executor runs it before returning and fixes line width, whitespace,
+  task checkboxes, and missing reports in the same session.
+- The current host performs review in `external_host` mode; the background
+  service must not spawn a nested supervisor CLI.
 - Treat `progress.jsonl`, the task contract, review results, and evidence as truth.
 - Executor cannot edit managed state or commit, push, publish, deploy, or bypass sandbox.
 - Supervisor cannot modify target files.
-- Non-zero exits, invalid JSON, missing session IDs, and review-time workspace drift block pass.
+- Invalid JSON, missing session IDs, and review-time workspace drift block pass. Absence checks use `assertion: negative` to express an expected exit 1/2 without guessing from command names or prose.
+- Expected-failure and fault-injection checks must use a parent command that
+  asserts the child's non-zero outcome and itself exits 0. Raw non-zero outcomes
+  stay in logs. Invalid negative evidence goes directly to Host review and never
+  triggers a full Executor invocation merely to rewrite JSON.
 - Engineering and SDD tasks need successful command evidence from at least two categories among build, test, and runtime/real invocation. Compile-only or unit-test-only delivery cannot enter formal review.
 - Run `superflow-managed-work-check.mjs <root> <task-id>` before delivery-ready.
+- `tasks.md` recognizes only local_required, environment_required, and release_required; untagged work is local. The Runner derives baseline evidence from the checklist, real workspace changes, test report, and successful commands. Executors add taskEvidence only for tasks that need distinct evidence.
+- The Runner merges the complete changedFiles set and stores structured command categories. The same mechanical rejection gets one automatic repair, with at most two automatic repairs overall, then the current host reviews it. Three or more permission denials in one invocation suppress another executor retry.
+- The Runner inherits valid startup, HTTP, test, and build commands across executor results. When the Host confirms complete historical evidence and only current structured metadata is invalid, it passes and promotes the artifact with an audit event instead of asking the Executor to rerun the environment or rewrite JSON.
+- When successful Spring Boot evidence already covers startup, real HTTP/runtime,
+  and evidence paths, and only the `invocation` category is missing, this is
+  verification-category ambiguity. Runner escalates directly to Host; after a
+  pass it promotes with an audit event instead of launching a full Executor for
+  metadata repair.
+- The Runner deterministically removes blank structured-array entries. A read-only
+  inspection with exit 1/2 that explicitly reports no process, listener,
+  container, or residue is a successful empty result; neither case launches the
+  Executor for metadata repair.
+- Status separates physical calls, infrastructure credits, and effective
+  development calls. Provider switches resume through
+  `--provider-switched <reason>`, and stage changes notify the Host conversation
+  without waking it for ordinary heartbeat.
+- Page/permission changes require frontend startup and Playwright/Cypress real-browser E2E. Cross-stack API changes require backend Controller and frontend request contracts plus an API-export snapshot.
 - Append the single `run.delivery_ready` event only after the integrity script passes. On failure, require human attention without leaving false delivery-ready evidence.
-- A passing run waits for explicit Git approval.
+- A passing run enters a three-stage source/environment/release state and still requires explicit Git approval.
+
+## Recovery
+
+After a background failure, recover from the task contract, event journal, run
+state, and latest checks. Old session IDs are audit records, not recovery
+dependencies. Plain `--resume-task` for a healthy `running` task is safe attach
+and continued wait only: it cannot reset state or budgets or launch another
+Executor. Only a recorded PID proven dead enters stale recovery; a missing PID
+fails closed into attach-and-wait. An active MCP wait's progress notifications
+are request-scoped, not autonomous callbacks after the Host turn ends.

package/assets/skills-en/superflow-pipeline/references/quality-gate.md

@@ -13,6 +13,16 @@
   evidence, necessity, simplest implementation, removed/rejected complexity,
   counts of new tables/fields/APIs/services/caches/async flows/jobs/compatibility
   layers, and a PASS/BLOCKED verdict. Only PASS may leave docs/design.
+- Full existing-system, DB/table, cross-repository, or real-entry changes have
+  `source-code-audit.md` with a Source Fact Freeze Card, all writers and real
+  callers, current/legacy/unmounted/data-model-only/owner-confirmed/blocked
+  classifications, DB check or skip reason, conflict audit, and owner boundary.
+- Database changes have `release-sql.md` with `database_change: true`, an exact
+  placeholder-free `target_sql_path`, `copy_policy: verbatim`, a valid
+  `sql_sha256`, and exactly one complete executable SQL block with required
+  prechecks, DDL/DML, migration, and post-verification. Design, tasks, and the
+  quality gate state that implementation agents may only copy it verbatim;
+  omissions return to docs.
 
 ## Design Gate
 
@@ -51,6 +61,8 @@
   sum(other components)` instead of calculating and rounding every component
   independently.
 - SQL risk review exists for DB changes.
+- Source-level design does not author or defer release SQL; it consumes the
+  already frozen `release-sql.md` contract.
 - No fallback or guesswork is allowed without explicit owner approval.
 
 ## Implementation Gate

package/assets/skills-en/superflow-pipeline/references/reviewer-checklist.md

@@ -22,6 +22,13 @@
   rounding explicit, without early rate rounding or synthesized inverses?
 - Are SQL changes reviewed and tied to total version SQL when needed?
 - Are RED/GREEN evidence and real-entry evidence present?
+- For L3/L4 negative controls, does the raw malformed or forbidden payload reach
+  the intended parser/validation branch, rather than merely returning the same
+  status code?
+- For task-owned resource cleanup, are every frozen identity field (nonce, PID,
+  port, signature, etc.) compared and mismatch refusal proven? An unchanged
+  certified helper needs integrity, task parameters, one failure cleanup, and
+  final zero residue only.
 - Are mock-only results clearly labeled?
 - Did workers stay inside allowed file boundaries?
 - Did hooks and guard scripts pass?

package/assets/skills-en/superflow-pipeline/scripts/superflow-guard.sh

@@ -322,6 +322,103 @@ change_has_field_status_risk() {
     "$CHANGE_DIR"/*.md "$CHANGE_DIR"/**/*.md 2>/dev/null
 }
 
+change_has_database_change() {
+  if grep -Eiq '^[[:space:]]*database:[[:space:]]*false([[:space:]]*#.*)?$' "$CHANGE_DIR/.openspec.yaml" 2>/dev/null; then
+    return 1
+  fi
+  if grep -Eiq '^[[:space:]]*database:[[:space:]]*true([[:space:]]*#.*)?$' "$CHANGE_DIR/.openspec.yaml" 2>/dev/null; then
+    return 0
+  fi
+  grep -RIEiq \
+    'database_change[[:space:]]*:[[:space:]]*true|database[ -]change.{0,12}(required|true|yes)|ALTER[[:space:]]+TABLE|CREATE[[:space:]]+TABLE|ADD[[:space:]]+(COLUMN|INDEX|KEY|CONSTRAINT)|schema[ -]migration|database[ -]migration|historical data.{0,12}(migration|backfill)|seed data.{0,12}(migration|backfill)|add.{0,12}(table|column|index|constraint)' \
+    "$CHANGE_DIR/proposal.md" \
+    "$CHANGE_DIR/api.md" \
+    "$CHANGE_DIR/design.md" \
+    "$CHANGE_DIR/tasks.md" \
+    "$CHANGE_DIR/database-contract.md" \
+    "$CHANGE_DIR/spec.md" \
+    "$CHANGE_DIR"/specs/*/spec.md 2>/dev/null
+}
+
+require_frozen_release_sql() {
+  local release_sql="$CHANGE_DIR/release-sql.md"
+  local output
+
+  require_file release-sql.md
+  [[ -f "$release_sql" ]] || return 0
+
+  output="$(mktemp)"
+  if ! python3 - "$release_sql" >"$output" 2>&1 <<'PY'
+import hashlib
+import re
+import sys
+from pathlib import Path
+
+path = Path(sys.argv[1])
+text = path.read_text(encoding="utf-8")
+errors = []
+
+front_match = re.match(r"\A---\n(.*?)\n---(?:\n|\Z)", text, re.S)
+front = front_match.group(1) if front_match else ""
+if not front_match:
+    errors.append("front matter is missing")
+
+def field(name):
+    match = re.search(rf"(?m)^{re.escape(name)}:\s*(.*?)\s*$", front)
+    return match.group(1).strip().strip('"\'') if match else ""
+
+if field("database_change").lower() != "true":
+    errors.append("database_change must be true")
+
+target = field("target_sql_path")
+if not target or not target.lower().endswith(".sql"):
+    errors.append("target_sql_path must be an exact .sql path")
+if re.search(r"[<>{}]|testN|version|pending|TBD", target, re.I):
+    errors.append("target_sql_path contains a placeholder")
+
+if field("copy_policy").lower() != "verbatim":
+    errors.append("copy_policy must be verbatim")
+
+declared_hash = field("sql_sha256").lower()
+if not re.fullmatch(r"[0-9a-f]{64}", declared_hash):
+    errors.append("sql_sha256 must be a 64-character lowercase SHA-256")
+
+blocks = re.findall(r"```sql[^\n]*\n(.*?)\n```", text, re.S | re.I)
+if len(blocks) != 1:
+    errors.append("release-sql.md must contain exactly one sql code block")
+else:
+    payload = blocks[0].replace("\r\n", "\n").replace("\r", "\n").rstrip("\n") + "\n"
+    actual_hash = hashlib.sha256(payload.encode("utf-8")).hexdigest()
+    if declared_hash and actual_hash != declared_hash:
+        errors.append(
+            f"sql_sha256 mismatch: declared={declared_hash} actual={actual_hash}"
+        )
+    executable = re.sub(r"/\*.*?\*/|--[^\n]*", "", payload, flags=re.S)
+    if not re.search(
+        r"\b(CREATE|ALTER|DROP|INSERT|UPDATE|DELETE|MERGE|TRUNCATE|CALL)\b",
+        executable,
+        re.I,
+    ):
+        errors.append("sql block has no executable schema/data change statement")
+    if not re.search(r"\b(SELECT|SHOW|CALL)\b", executable, re.I):
+        errors.append("sql block has no executable verification statement")
+
+if errors:
+    print("; ".join(errors))
+    sys.exit(1)
+PY
+  then
+    issues+=("frozen release SQL invalid: $(tr '\n' ' ' < "$output" | sed -E 's/[[:space:]]+/ /g')")
+  fi
+  rm -f "$output"
+
+  require_grep 'release-sql\.md' design.md "frozen release SQL design reference"
+  require_grep 'release-sql\.md' tasks.md "frozen release SQL task reference"
+  require_grep 'release-sql\.md' sdd-quality-gate.md "frozen release SQL quality-gate reference"
+  require_grep 'verbatim' tasks.md "verbatim SQL copy task"
+  require_grep 'verbatim' sdd-quality-gate.md "verbatim SQL copy quality gate"
+}
+
 change_contract_docs_without_complexity_budget() {
   local path
   for path in proposal.md api.md design.md tests.md spec.md; do
@@ -340,6 +437,12 @@ change_all_docs_without_complexity_budget() {
 }
 
 change_has_external_config_risk() {
+  if grep -Eiq '^[[:space:]]*external_config:[[:space:]]*false([[:space:]]*#.*)?$' "$CHANGE_DIR/.openspec.yaml" 2>/dev/null; then
+    return 1
+  fi
+  if grep -Eiq '^[[:space:]]*external_config:[[:space:]]*true([[:space:]]*#.*)?$' "$CHANGE_DIR/.openspec.yaml" 2>/dev/null; then
+    return 0
+  fi
   local content
   content="$(change_contract_docs_without_complexity_budget)"
   grep -Eiq \
@@ -348,6 +451,12 @@ change_has_external_config_risk() {
 }
 
 change_has_concurrency_idempotency_risk() {
+  if grep -Eiq '^[[:space:]]*concurrency:[[:space:]]*false([[:space:]]*#.*)?$' "$CHANGE_DIR/.openspec.yaml" 2>/dev/null; then
+    return 1
+  fi
+  if grep -Eiq '^[[:space:]]*concurrency:[[:space:]]*true([[:space:]]*#.*)?$' "$CHANGE_DIR/.openspec.yaml" 2>/dev/null; then
+    return 0
+  fi
   grep -RIEiq \
     'concurren|idempoten|race condition|duplicate (request|callback|consumption|delivery)|atomic claim|batch (issue|activate|renew)' \
     "$CHANGE_DIR/proposal.md" \
@@ -359,6 +468,12 @@ change_has_concurrency_idempotency_risk() {
 }
 
 change_has_money_precision_risk() {
+  if grep -Eiq '^[[:space:]]*money:[[:space:]]*false([[:space:]]*#.*)?$' "$CHANGE_DIR/.openspec.yaml" 2>/dev/null; then
+    return 1
+  fi
+  if grep -Eiq '^[[:space:]]*money:[[:space:]]*true([[:space:]]*#.*)?$' "$CHANGE_DIR/.openspec.yaml" 2>/dev/null; then
+    return 0
+  fi
   grep -RIEiq \
     '(^|[^[:alnum:]_])(amount|fee|fees|price|discount|deduction|refund|payment|invoice|balance|proration|allocation|reconciliation|residual|precision|rounding)([^[:alnum:]_]|$)|revenue sharing|profit sharing|split payment|electricity fee|service fee|package settlement|serviceFee|chargeFee|totalAmount|actualAmount|payAmount|refundAmount|discountAmount|invoiceAmount|balanceAmount' \
     "$CHANGE_DIR/proposal.md" \
@@ -369,6 +484,53 @@ change_has_money_precision_risk() {
     "$CHANGE_DIR"/specs/*/spec.md 2>/dev/null
 }
 
+change_has_source_fact_risk() {
+  grep -RIEiq \
+    'existing|current behavior|legacy|database|table relationship|Mapper|XML|cross[- ](repo|service)|crosses services|real entry|mini[- ]program|H5' \
+    "$CHANGE_DIR/proposal.md" \
+    "$CHANGE_DIR/api.md" \
+    "$CHANGE_DIR/design.md" \
+    "$CHANGE_DIR/tests.md" \
+    "$CHANGE_DIR/spec.md" \
+    "$CHANGE_DIR"/specs/*/spec.md 2>/dev/null
+}
+
+change_has_cardinality_conflict_signal() {
+  grep -RIEiq \
+    'List[<(]|orderIds|batchInsert|one[- ]to[- ]many|relation table' \
+    "$CHANGE_DIR"/*.md "$CHANGE_DIR"/**/*.md 2>/dev/null
+}
+
+require_source_fact_audit() {
+  require_file source-code-audit.md
+  require_grep 'Source Fact Freeze Card' source-code-audit.md "source fact freeze card"
+  require_grep 'Business conclusion' source-code-audit.md "source fact business conclusion"
+  require_grep 'understand.*locator' source-code-audit.md "understand-anything locator evidence"
+  require_grep 'Data model' source-code-audit.md "source fact data model"
+  require_grep 'All writers' source-code-audit.md "all writers inventory"
+  require_grep 'Real user entr' source-code-audit.md "real user entry"
+  require_grep 'Current callers' source-code-audit.md "current callers"
+  require_grep 'Legacy conflict' source-code-audit.md "legacy conflict"
+  require_grep 'DB (check|required|skip|reason)' source-code-audit.md "DB check or skip reason"
+  require_grep 'Conclusion level' source-code-audit.md "conclusion level"
+  require_grep 'owner.*decision' source-code-audit.md "owner decision boundary"
+  require_grep 'current' source-code-audit.md "current evidence classification"
+  require_grep 'legacy' source-code-audit.md "legacy evidence classification"
+  require_grep 'unmounted' source-code-audit.md "unmounted evidence classification"
+  require_grep 'data-model-only' source-code-audit.md "data-model-only evidence classification"
+  require_grep 'owner-confirmed' source-code-audit.md "owner-confirmed evidence classification"
+  require_grep 'blocked' source-code-audit.md "blocked evidence classification"
+  require_grep 'Evidence classifications.*current.*legacy.*unmounted.*data-model-only.*owner-confirmed.*blocked' source-code-audit.md "complete evidence classification set"
+  require_grep 'Question Eligibility Gate' source-code-audit.md "question eligibility gate"
+  require_grep 'Source search' source-code-audit.md "source search evidence"
+  require_grep 'Mapper.*SQL|SQL.*Mapper' source-code-audit.md "Mapper and SQL search evidence"
+  require_grep 'Frontend.*mini[- ]program.*H5' source-code-audit.md "frontend mini-program H5 caller search"
+  require_grep 'sibling.*repo' source-code-audit.md "sibling repository search"
+  if change_has_cardinality_conflict_signal; then
+    require_grep 'Conflict Audit' source-code-audit.md "cardinality conflict audit"
+  fi
+}
+
 change_has_fx_precision_risk() {
   grep -RIEiq \
     '(^|[^[:alnum:]_])(exchange rate|fx rate|currency conversion|cross.currency|base currency|quote currency)([^[:alnum:]_]|$)' \
@@ -388,8 +550,10 @@ change_has_architecture_boundary_risk() {
     <<< "$content"
 }
 
-first_prompt_rel() {
-  find "$CHANGE_DIR" -path '*/prompt/*.md' -type f | head -n 1 | sed "s#^$CHANGE_DIR/##"
+prompt_rels() {
+  find "$CHANGE_DIR" -path '*/prompt/*.md' -type f -print \
+    | LC_ALL=C sort \
+    | sed "s#^$CHANGE_DIR/##"
 }
 
 require_any_prompt() {
@@ -399,7 +563,7 @@ require_any_prompt() {
 }
 
 require_prompt_set() {
-  local task_count prompt_count non_index_prompt_count
+  local task_count prompt_count non_index_prompt_count prompt_rel document
   require_file prompt/implementation.md
   task_count="$(grep -E '^[[:space:]]*-[[:space:]]+\[[ xX]\]' "$CHANGE_DIR/tasks.md" 2>/dev/null | wc -l | tr -d ' ')"
   prompt_count="$(find "$CHANGE_DIR/prompt" -maxdepth 1 -type f -name '*.md' 2>/dev/null | wc -l | tr -d ' ')"
@@ -410,9 +574,15 @@ require_prompt_set() {
   if [[ "$prompt_count" -eq 0 ]]; then
     issues+=("missing prompt/*.md files")
   fi
-  if [[ -f "$CHANGE_DIR/tasks.md" ]] && ! grep -Riq 'prompt/.*\.md' "$CHANGE_DIR/tasks.md" "$CHANGE_DIR/traceability-matrix.md" "$CHANGE_DIR/sdd-quality-gate.md" "$CHANGE_DIR/test-report.md" 2>/dev/null; then
-    issues+=("prompt files are not cross-linked from tasks/traceability/quality gate/test-report")
-  fi
+  for document in tasks.md traceability-matrix.md sdd-quality-gate.md test-report.md; do
+    [[ -f "$CHANGE_DIR/$document" ]] || continue
+    while IFS= read -r prompt_rel; do
+      if ! grep -Fq "]($prompt_rel)" "$CHANGE_DIR/$document" \
+        && ! grep -Fq "](./$prompt_rel)" "$CHANGE_DIR/$document"; then
+        issues+=("$document does not link $prompt_rel")
+      fi
+    done < <(prompt_rels)
+  done
 }
 
 require_money_precision_contract() {
@@ -548,11 +718,27 @@ case "$PHASE" in
     require_grep 'RED|red-green|GREEN' tests.md "RED/GREEN test contract"
     require_grep 'curl|Postman|Newman|pytest|RestAssured|automation command' tests.md "interface automation command"
     require_grep 'handoff_hash|sdd-context|context package|anti-drift' sdd-quality-gate.md "handoff/context-drift gate"
+    if change_has_database_change; then
+      require_frozen_release_sql
+    fi
     workflow="$(state_get workflow)"
     if [[ "$workflow" == "full" ]]; then
+      require_file requirement-review.md
+      require_grep '^review_verdict:[[:space:]]*PASS[[:space:]]*$' requirement-review.md "requirement reverse review PASS verdict"
+      require_grep '^implementation_readiness:[[:space:]]*READY[[:space:]]*$' requirement-review.md "requirement review implementation readiness"
+      require_grep '^open_blockers:[[:space:]]*0[[:space:]]*$' requirement-review.md "requirement reverse review has no open blockers"
+      require_grep '^open_external_contracts:[[:space:]]*0[[:space:]]*$' requirement-review.md "requirement review has no open external contracts"
+      require_grep '^open_source_investigations:[[:space:]]*0[[:space:]]*$' requirement-review.md "requirement review has no open source investigations"
+      require_grep '^open_owner_decisions:[[:space:]]*0[[:space:]]*$' requirement-review.md "requirement review has no open owner decisions"
+      require_grep 'Closure Matrix' requirement-review.md "requirement review closure matrix"
+      require_grep 'Review Findings' requirement-review.md "requirement review findings"
+      require_grep 'Open Blockers' requirement-review.md "requirement review open blocker ledger"
       require_minimal_design_review design.md "complexity reduction review"
       require_grep 'Minimal Design Review|Complexity Reduction Review' sdd-quality-gate.md "complexity reduction quality gate"
       require_minimal_design_pass sdd-quality-gate.md
+      if change_has_source_fact_risk; then
+        require_source_fact_audit
+      fi
     fi
     if change_has_money_precision_risk; then
       require_grep 'Money Precision Boundary' sdd-quality-gate.md "money precision quality gate"
@@ -583,6 +769,9 @@ case "$PHASE" in
     require_handoff_current
     require_beta_handoff_structure
     require_hash_recorded
+    if change_has_database_change; then
+      require_frozen_release_sql
+    fi
     workflow="$(state_get workflow)"
     if [[ "$workflow" == "full" ]]; then
       require_minimal_design_review design.md "complexity reduction review"
@@ -622,11 +811,14 @@ case "$PHASE" in
     require_handoff_current
     require_beta_handoff_structure
     require_hash_recorded
+    if change_has_database_change; then
+      require_frozen_release_sql
+    fi
     require_any_prompt
     require_prompt_set
     require_grep 'OpenSpec/SDD|canonical design source|canonical|source of truth' design.md "canonical source boundary"
-    prompt_rel="$(first_prompt_rel)"
-    if [[ -n "${prompt_rel:-}" ]]; then
+    while IFS= read -r prompt_rel; do
+      [[ -n "${prompt_rel:-}" ]] || continue
       require_grep 'Superpowers Technical Design inheritance|Superpowers execution strategy inheritance|technical_design|source-level HOW' "$prompt_rel" "prompt Superpower strategy inheritance"
       technical_design_rel="$(state_get technical_design)"
       if [[ -n "${technical_design_rel:-}" && "$technical_design_rel" != "null" ]]; then
@@ -645,7 +837,11 @@ case "$PHASE" in
         fi
       fi
       require_grep 'context anti-drift|handoff_hash|sdd-context' "$prompt_rel" "prompt context drift inheritance"
-    fi
+      if change_has_database_change; then
+        require_grep 'release-sql\.md' "$prompt_rel" "prompt frozen release SQL reference"
+        require_grep 'verbatim|sql_sha256' "$prompt_rel" "prompt verbatim SQL copy contract"
+      fi
+    done < <(prompt_rels)
     require_state_value build_mode "build_mode"
     require_state_value isolation "isolation"
     require_state_value tdd_mode "tdd_mode"

package/assets/skills-en/superflow-pipeline/scripts/superflow-handoff.sh

@@ -244,10 +244,13 @@ sync_state() {
   "$state_sh" set "$CHANGE_DIR" context_compression "$context_compression"
   "$state_sh" set "$CHANGE_DIR" handoff_context .sdd/handoff/sdd-context.json
   "$state_sh" set "$CHANGE_DIR" handoff_hash "$context_hash"
-  technical_design_rel="$(
-    find "$CHANGE_DIR/docs/superpowers/specs" -maxdepth 1 -type f -name '*.md' 2>/dev/null \
-      | sort | head -n 1 | sed "s#^$CHANGE_DIR/##"
-  )"
+  technical_design_rel=""
+  if [[ -d "$CHANGE_DIR/docs/superpowers/specs" ]]; then
+    technical_design_rel="$(
+      find "$CHANGE_DIR/docs/superpowers/specs" -maxdepth 1 -type f -name '*.md' \
+        | sort | head -n 1 | sed "s#^$CHANGE_DIR/##"
+    )"
+  fi
   if [[ -n "$technical_design_rel" ]]; then
     "$state_sh" set "$CHANGE_DIR" technical_design "$technical_design_rel"
   fi