@mmerterden/multi-agent-pipeline 16.11.0 → 16.12.0

package/pipeline/schemas/analysis-spec.schema.json

@@ -24,6 +24,8 @@
         "awaiting_pass_b_approval",
         "rendering_pass_b",
         "drafting",
+        "reviewing_draft",
+        "reviewing_gaps",
         "awaiting_output_decision",
         "dispatching",
         "reporting",

package/pipeline/schemas/reviewer-output.schema.json

@@ -3,7 +3,7 @@
   "$id": "https://github.com/mmerterden/multi-agent-pipeline/pipeline/schemas/reviewer-output.schema.json",
   "version": "1.1.0",
   "title": "Multi-Agent Pipeline  -  Phase 4 reviewer output",
-  "description": "Contract for a single code-reviewer subagent's JSON output in Phase 4 Step 2. Reviewer set is CLI-aware: Claude Code dispatches 2 parallel reviewers (Fable, Sonnet); Copilot CLI dispatches 3 (Opus, GPT-5.4, Sonnet); Codex CLI dispatches 3 (gpt-5.6 at xhigh, gpt-5.4, gpt-5.6 at medium). Every reviewer must return an object matching this shape before Opus triage merges them. v1.1.0 adds the rule-ID conformance checklist: when the orchestrator supplies a ${CRITERIA} block (Phase 4 Step 1.78), the reviewer must return one conformance row per selected rule ID. Findings alone cannot answer 'was this applied completely' - a reviewer that opened nothing returns the same empty findings array as one that checked everything.",
+  "description": "Contract for a single code-reviewer subagent's JSON output in Phase 4 Step 2. Every host dispatches 3 parallel reviewers; the middle slot is CLI-aware: Claude Code (Fable, Opus, Sonnet); Copilot CLI (Opus, GPT-5.4, Sonnet); Codex CLI dispatches 3 (gpt-5.6 at xhigh, gpt-5.4, gpt-5.6 at medium). Every reviewer must return an object matching this shape before Opus triage merges them. v1.1.0 adds the rule-ID conformance checklist: when the orchestrator supplies a ${CRITERIA} block (Phase 4 Step 1.78), the reviewer must return one conformance row per selected rule ID. Findings alone cannot answer 'was this applied completely' - a reviewer that opened nothing returns the same empty findings array as one that checked everything.",
   "type": "object",
   "additionalProperties": false,
   "required": ["findings", "approved"],

package/pipeline/schemas/triage-output.schema.json

@@ -102,7 +102,7 @@
         "reviewerCount": {
           "type": "integer",
           "minimum": 1,
-          "description": "Reviewers dispatched this iteration (Claude Code = 2, Copilot CLI = 3, Codex CLI = 3)."
+          "description": "Reviewers dispatched this iteration (Claude Code = 3, Copilot CLI = 3, Codex CLI = 3)."
         },
         "verdict": {
           "type": "string",

package/pipeline/scripts/anonymize-findings.mjs

@@ -1,7 +1,7 @@
 #!/usr/bin/env node
 // anonymize-findings.mjs  -  strip reviewer identity before a model judges findings.
 //
-// Phase 4 dispatches 2 reviewers on Claude Code (Fable + Sonnet) and 3 on
+// Phase 4 dispatches 3 reviewers on Claude Code (Fable + Opus + Sonnet) and 3 on
 // Copilot CLI (GPT + Opus + Sonnet), then hands the merged findings to triage -
 // Fable on Claude Code, Opus on Copilot. On both sides the triage model is also
 // one of the reviewers, and Step 3.2 used to label the input "Reviewer 1 +

package/pipeline/scripts/smoke-cross-cli-behavior.sh

@@ -185,12 +185,12 @@ else
 fi
 
 # ──────────────────────────────────────────────────────────────────────────
-# CLI-aware reviewer-count contract: Claude Code = 2 (Fable + Sonnet), Copilot CLI
+# CLI-aware reviewer-count contract: Claude Code = 3 (Fable + Opus + Sonnet), Copilot CLI
 # = 3 (Opus + GPT-5.4 + Sonnet), Codex CLI = 3 (gpt-5.6 xhigh + gpt-5.4 + gpt-5.6
 # medium). Nothing in code enforces the count (the orchestrator dispatches per the
 # doc), so lock the CONTRACT here against drift across the phase doc, the schema,
 # and the consensus block.
-echo "→ reviewer-count contract (Claude=2, Copilot=3, Codex=3)"
+echo "→ reviewer-count contract (Claude=3, Copilot=3, Codex=3)"
 # `/multi-agent:update` runs this smoke on the user's machine, where the tree is
 # ~/.claude/{schemas,multi-agent-refs} and NOT <root>/pipeline/*. Resolving the
 # root by hand here reported three phantom failures in an install, which reads as
@@ -203,10 +203,10 @@ TRSCHEMA="${MA_SCHEMAS:+$MA_SCHEMAS/triage-output.schema.json}"
 
 if [ -z "$P4" ] || [ ! -f "$P4" ]; then
   echo "  ↷ SKIP: phase-4-review.md not present in this $MA_LAYOUT layout"
-elif grep -qiE "Claude Code (dispatches|=|:) ?2|2-model" "$P4" \
+elif grep -qiE "Claude Code (dispatches|=|:) ?3|Claude Code dispatches Fable \+ Opus \+ Sonnet" "$P4" \
   && grep -qiE "Copilot CLI (dispatches|=|:) ?3|3-model" "$P4" \
-  && grep -qiE "Codex CLI (dispatches|=|:) ?3|Claude Code 2, Copilot CLI 3, Codex CLI 3" "$P4"; then
-  pass "phase-4-review declares Claude=2 / Copilot=3 / Codex=3 reviewers"
+  && grep -qiE "Codex CLI (dispatches|=|:) ?3|Claude Code 3, Copilot CLI 3, Codex CLI 3" "$P4"; then
+  pass "phase-4-review declares Claude=3 / Copilot=3 / Codex=3 reviewers"
 else
   fail "phase-4-review does not declare the CLI-aware reviewer count for all three hosts"
 fi
@@ -231,13 +231,13 @@ else
   fail "reviewer-output schema missing the CLI-aware note"
 fi
 
-# consensus.reviewerCount must accommodate both 2 and 3 (min 1, no max < 3)
+# consensus.reviewerCount must accommodate 3 on every host (min 1, no max < 3)
 if [ -z "$TRSCHEMA" ] || [ ! -f "$TRSCHEMA" ]; then
   echo "  ↷ SKIP: triage-output.schema.json not present in this $MA_LAYOUT layout"
-elif grep -q '"reviewerCount"' "$TRSCHEMA" && grep -qi "Claude Code = 2, Copilot CLI = 3" "$TRSCHEMA"; then
-  pass "triage consensus.reviewerCount documents 2 (Claude) and 3 (Copilot)"
+elif grep -q '"reviewerCount"' "$TRSCHEMA" && grep -qi "Claude Code = 3, Copilot CLI = 3" "$TRSCHEMA"; then
+  pass "triage consensus.reviewerCount documents 3 reviewers on every host"
 else
-  fail "triage consensus.reviewerCount does not document the 2/3 split"
+  fail "triage consensus.reviewerCount does not document the 3-reviewer set"
 fi
 
 echo ""

package/pipeline/scripts/validate-analysis-doc.mjs

@@ -215,11 +215,15 @@ function main() {
     }
   }
 
-  // 2a. Corporate profile: every EKLENECEK owes an open question.
+  // 2a. Corporate profile: every EKLENECEK owes an open question, by id.
   //
   // A placeholder with nobody assigned to resolve it is how "to be filled in"
-  // reaches production. The pairing is what makes the backbone guarantee worth
-  // having: the section renders, AND the gap it admits to is tracked.
+  // reaches production. This used to pair a gap with Section 20 by matching the
+  // section NUMBER inside the risks block, which is ambiguous in both directions:
+  // "9" is a substring of "19" and "5.9" of "5.9.5", so the regex grew guards for
+  // cases it still could not separate. Stable ids remove the ambiguity the same
+  // way IG / UC / FG already do: the body writes EKLENECEK (AS-07) and Section 20
+  // defines AS-07, and each side is checked against the other.
   if (profile === "corporate") {
     const lines = text.split("\n");
     const risksStart = lines.findIndex((l) =>
@@ -232,29 +236,61 @@ function main() {
         risksBlock += `${lines[i]}\n`;
       }
     }
-    let currentSection = null;
-    const unpaired = new Set();
+    const AS_ID = /\bAS-(\d{2,3})\b/g;
+    const definedIds = new Set(
+      [...risksBlock.matchAll(AS_ID)].map((m) => `AS-${m[1]}`),
+    );
+    const referencedIds = new Set();
     for (let i = 0; i < lines.length; i++) {
-      const m = lines[i].match(/^#{2,4}\s+(\d+(?:\.\d+)*)\.?\s/);
-      if (m) currentSection = m[1];
-      if (!lines[i].includes("EKLENECEK")) continue;
       if (risksStart !== -1 && i > risksStart) continue;
-      if (!currentSection) continue;
-      // The number has to stand alone: substring matching would let section "2"
-      // be satisfied by a "2.3" in some other row, and "9" by a "19". The
-      // trailing guard rejects a following digit and a following ".<digit>",
-      // but NOT a sentence-ending period - "etkilediği bölüm 9." is a real
-      // pairing, and rejecting it would block a correct document.
-      const token = new RegExp(
-        `(^|[^\\d.])${currentSection.replace(/\./g, "\\.")}(?!\\d|\\.\\d)`,
-        "m",
-      );
-      if (!token.test(risksBlock)) unpaired.add(currentSection);
+      for (const m of lines[i].matchAll(AS_ID)) referencedIds.add(`AS-${m[1]}`);
+      if (!lines[i].includes("EKLENECEK")) continue;
+      // A gap that names no id cannot be traced to an owner, which is the whole
+      // point of admitting it.
+      if (!/\bAS-\d{2,3}\b/.test(lines[i])) {
+        errors.push(
+          `line ${i + 1}: EKLENECEK carries no AS-NN reference (Locked 33)`,
+        );
+      }
     }
-    for (const sec of unpaired) {
-      errors.push(
-        `section ${sec} carries EKLENECEK but no Risks and Open Questions row references it (Locked 33)`,
-      );
+    for (const id of referencedIds) {
+      if (!definedIds.has(id)) {
+        errors.push(
+          `${id} is referenced in the body but not defined in Risks and Open Questions (Locked 33)`,
+        );
+      }
+    }
+    for (const id of definedIds) {
+      if (!referencedIds.has(id)) {
+        // An open question with no body reference is allowed only when the row
+        // says so; otherwise it is a question about nothing the document raises.
+        const row = risksBlock
+          .split("\n")
+          .find((l) => l.includes(id)) || "";
+        if (!/(no body reference|govde referansi yok)/i.test(row)) {
+          errors.push(
+            `${id} is defined in Risks and Open Questions but never referenced in the body (Locked 33)`,
+          );
+        }
+      }
+    }
+
+    // 2b. JSON does not fit in a table cell (Locked 37).
+    //
+    // One document rendered `Request` as inline JSON with a field note glued to
+    // the end, and `Response` as a proper fenced block below the table: two
+    // formats for the same thing, one table apart. The payload belongs under the
+    // table; the cell says which of the three states it is in.
+    for (let i = 0; i < lines.length; i++) {
+      const line = lines[i];
+      if (!/^\|\s*\*\*(Request|Response)\*\*\s*\|/.test(line)) continue;
+      const cell = line.split("|")[2] || "";
+      if (/[{}]/.test(cell)) {
+        errors.push(
+          `line ${i + 1}: ${/Request/.test(line) ? "Request" : "Response"} cell holds JSON; ` +
+            "put the payload in a fenced block under the table (Locked 37)",
+        );
+      }
     }
 
     // The traceability matrix is the most expensive thing in the document to get
@@ -295,6 +331,32 @@ function main() {
     }
   }
 
+  // 2c. Diagram labels keep their diacritics.
+  //
+  // The punctuation gate skips fenced blocks, so a mermaid diagram is the one
+  // place a Turkish document can quietly go ASCII: one run shipped `Hayir` and
+  // `Basarili` in its flow chart while every other line was correct. A warning,
+  // not an error - node ids are legitimately ASCII, only labels are checked.
+  if (parsed.fm.language === "tr") {
+    const mmLines = text.split("\n");
+    let inMermaid = false;
+    for (let i = 0; i < mmLines.length; i++) {
+      const t = mmLines[i].trim();
+      if (/^```mermaid\s*$/.test(t)) { inMermaid = true; continue; }
+      if (inMermaid && t === "```") { inMermaid = false; continue; }
+      if (!inMermaid) continue;
+      const labels = [...mmLines[i].matchAll(/[[({|]([^\]})|]{3,})[\]})|]/g)].map((m) => m[1]);
+      for (const lbl of labels) {
+        if (/^[\x20-\x7E]+$/.test(lbl) && /\b(Hayir|Basarili|Basarisiz|Odeme|Iptal|Gecerli|Gecersiz|Secim|Dogrulama|Uyari|Baslangic|Bitis|Onay|Aciklama)\b/.test(lbl)) {
+          errors.push(
+            `WARN line ${i + 1}: diagram label "${lbl}" looks like Turkish flattened to ASCII (Locked 7)`,
+          );
+        }
+      }
+    }
+  }
+
+
   // 2b. Opt-in coverage sections must be present when the front-matter says so.
   const uiTests = String(parsed?.fm?.ui_tests || "false").toLowerCase() === "true";
   const a11yDepth = String(parsed?.fm?.a11y_depth || "basic").toLowerCase();

package/pipeline/skills/.skills-index.json

@@ -950,7 +950,7 @@
     },
     {
       "name": "multi-agent",
-      "description": "Task orchestrator: runs the full pipeline from a Jira ID or GitHub Issue URL  -  analysis → plan → TDD development → parallel review (Fable + Sonnet on Claude Code, GPT + Opus + Sonnet on Copilot CLI) → commit → log. Every step is written to agent-log.md. Use when given a Jira ID, a GitHub issue or ",
+      "description": "Task orchestrator: runs the full pipeline from a Jira ID or GitHub Issue URL  -  analysis → plan → TDD development → parallel review (Fable + Opus + Sonnet on Claude Code, GPT + Opus + Sonnet on Copilot CLI) → commit → log. Every step is written to agent-log.md. Use when given a Jira ID, a GitHub issue or ",
       "platform": null,
       "group": "core",
       "plugin": null,
@@ -1313,7 +1313,7 @@
     },
     {
       "name": "multi-agent-review",
-      "description": "Run parallel review on a branch diff or a Pull Request: 2 models on Claude Code (Fable + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonnet). On PR input, posts per-finding inline comments + approve/needs-work. With no input (interactive), lists open GitHub + Bitbucket PRs to multi-select. Use wh",
+      "description": "Run parallel review on a branch diff or a Pull Request: 3 models on Claude Code (Fable + Opus + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonnet). On PR input, posts per-finding inline comments + approve/needs-work. With no input (interactive), lists open GitHub + Bitbucket PRs to multi-select. Use wh",
       "platform": null,
       "group": "core",
       "plugin": null,

package/pipeline/skills/shared/README.md

@@ -60,7 +60,7 @@ Source layout is logical grouping only  -  skill discovery at runtime is unchang
 | [`multi-agent-refactor`](./core/multi-agent-refactor/) | `core` | Analyse the project: extract adapted best-practices, hunt real bugs + improvement areas, check upstream drift of derived skills, research th |
 | [`multi-agent-resume`](./core/multi-agent-resume/) | `core` | Resume a stopped or failed task from the phase where it left off. Use when a task stopped or failed and should carry on from where it left o |
 | [`multi-agent-resume-local`](./core/multi-agent-resume-local/) | `core` | Continue already-done LOCAL work through the pipeline tail: Review → Build+Test → Commit/PR → Report (technical analysis + Jira test-scenari |
-| [`multi-agent-review`](./core/multi-agent-review/) | `core` | Run parallel review on a branch diff or a Pull Request: 2 models on Claude Code (Fable + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonn |
+| [`multi-agent-review`](./core/multi-agent-review/) | `core` | Run parallel review on a branch diff or a Pull Request: 3 models on Claude Code (Fable + Opus + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonn |
 | [`multi-agent-review-analysis`](./core/multi-agent-review-analysis/) | `core` | Review a written analysis document instead of a diff: resolve it from a path, a Confluence page or a Jira issue, run the deterministic gates |
 | [`multi-agent-review-issue`](./core/multi-agent-review-issue/) | `core` | Assess whether a GitHub issue is ready for multi-agent development: fetch it, grade scope / acceptance criteria / repro / design / API / sta |
 | [`multi-agent-review-jira`](./core/multi-agent-review-jira/) | `core` | Assess whether a Jira issue is ready for multi-agent development: fetch it, grade scope / acceptance criteria / repro / design / API / stack |

package/pipeline/skills/shared/core/multi-agent/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: multi-agent
 language: en
-description: "Task orchestrator: runs the full pipeline from a Jira ID or GitHub Issue URL  -  analysis → plan → TDD development → parallel review (Fable + Sonnet on Claude Code, GPT + Opus + Sonnet on Copilot CLI) → commit → log. Every step is written to agent-log.md. Use when given a Jira ID, a GitHub issue or a free-text task and the whole pipeline should run."
+description: "Task orchestrator: runs the full pipeline from a Jira ID or GitHub Issue URL  -  analysis → plan → TDD development → parallel review (Fable + Opus + Sonnet on Claude Code, GPT + Opus + Sonnet on Copilot CLI) → commit → log. Every step is written to agent-log.md. Use when given a Jira ID, a GitHub issue or a free-text task and the whole pipeline should run."
 user-invocable: true
 argument-hint: '"PROJ-12345" "feature/PROJ-12345-flight-filter" | "https://github.com/.../issues/316" | status | log #1 | resume #1 | kill #1 | clear-logs | purge | review'
 ---
@@ -455,7 +455,7 @@ For each todo (respecting dependency order):
 ### Phase 4: Review (parallel + triage)
 0. **Diff Risk Scoring (advisory, v8.3+)**  -  before reviewer dispatch run `node $HOME/.claude/scripts/diff-risk-score.mjs --base "$BASE_BRANCH" --top 5` and inject the top-N risk-ranked files as a `${PRIORITY_FILES}` block into each reviewer's prompt. Heuristic, deterministic, sub-second, never gates the pipeline. Disabled when `prefs.global.diffRiskAdvisory = false`. Signals: security paths (×3), schema migrations (×4), public API surfaces (×2), no-test-change (×2.5), complexity delta (×1.5), UI-critical paths (×1.5), loc changed (×1).
 1. Launch **code-reviewer** agents in parallel. Reviewer set depends on which CLI is hosting the pipeline:
-   - **Claude Code** (2 reviewers): `claude-fable-5` (deep security + architecture) + `claude-sonnet-5` (quality + correctness)
+   - **Claude Code** (3 reviewers): `claude-fable-5` (deep security + architecture) + `claude-sonnet-5` (quality + correctness)
    - **Copilot CLI** (3 reviewers): `gpt-5.4` (edge cases, different perspective) + `claude-opus-5` + `claude-sonnet-5` (Fable 5 is not offered on Copilot CLI)
    - Triage: single top-tier pass over merged findings (`claude-fable-5` on Claude Code, `claude-opus-5` on Copilot CLI)
 2. Collect findings, classify:
@@ -656,7 +656,7 @@ Phase 3: Dev
        └─ TDD, token rules, code style  -  from global instructions
 
 Phase 4: Review (parallel + triage)
-  Claude Code  (2 parallel):
+  Claude Code  (3 parallel):
     ├─ code-reviewer (claude-opus)   → all 5 review skills
     └─ code-reviewer (claude-sonnet) → all 5 review skills
   Copilot CLI  (3 parallel):

package/pipeline/skills/shared/core/multi-agent-help/SKILL.md

@@ -63,7 +63,7 @@ Pipeline (after Phase 0):
   Phase 2: Planning  -> Task breakdown + architecture review + Plan Approval Gate
   Phase 3: Dev       -> TDD: test -> code -> build (Sonnet) + build queue
   Phase 4: Review    -> Deterministic gates + parallel AI review + Fable triage
-                        (Claude Code: Fable + Sonnet · Copilot CLI: GPT-5.4 + Opus + Sonnet)
+                        (Claude Code: Fable + Opus + Sonnet · Copilot CLI: GPT-5.4 + Opus + Sonnet)
   Phase 5: Test      -> Optional: switch to branch, test in Xcode
   Phase 6: Commit    -> Commit -> push -> PR + issue body update (never auto-closes)
   Phase 7: Report    -> Channels dispatcher (PR · Jira · Confluence · Wiki, multi-select)
@@ -240,7 +240,7 @@ Pipeline (Phase 0'dan sonra):
   Phase 2: Planning  -> Task kırılımı + mimari inceleme + Plan Onay Kapısı
   Phase 3: Dev       -> TDD: test -> kod -> build (Sonnet) + build queue
   Phase 4: Review    -> Deterministik kapılar + paralel AI review + Fable triage
-                        (Claude Code: Fable + Sonnet · Copilot CLI: GPT-5.4 + Opus + Sonnet)
+                        (Claude Code: Fable + Opus + Sonnet · Copilot CLI: GPT-5.4 + Opus + Sonnet)
   Phase 5: Test      -> Opsiyonel: branch'e geç, Xcode'da test
   Phase 6: Commit    -> Commit -> push -> PR + issue body güncelleme (hiç auto-close yok)
   Phase 7: Report    -> Channels dispatcher (PR · Jira · Confluence · Wiki, multi-select)

package/pipeline/skills/shared/core/multi-agent-review/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: multi-agent-review
 language: en
-description: "Run parallel review on a branch diff or a Pull Request: 2 models on Claude Code (Fable + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonnet). On PR input, posts per-finding inline comments + approve/needs-work. With no input (interactive), lists open GitHub + Bitbucket PRs to multi-select. Use when a branch diff or a pull request needs reviewing before it merges."
+description: "Run parallel review on a branch diff or a Pull Request: 3 models on Claude Code (Fable + Opus + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonnet). On PR input, posts per-finding inline comments + approve/needs-work. With no input (interactive), lists open GitHub + Bitbucket PRs to multi-select. Use when a branch diff or a pull request needs reviewing before it merges."
 user-invocable: true
 argument-hint: "[#N | repo#N | PR-URL | branch]  -  PR by number/URL, repo+number, or branch (GitHub + Bitbucket Server). If omitted: interactive PR picker, else current branch."
 ---
@@ -33,7 +33,7 @@ Skip Phase 0-3 and review a diff only. Input shapes: a PR (`#N`, `repo#N`, GitHu
 
 3. **Start parallel reviewers**  -  reviewer set depends on the host CLI:
 
-   **Claude Code (2 in parallel):**
+   **Claude Code (3 in parallel):**
    - Agent 1: `claude-fable-5` → security + architecture
    - Agent 2: `claude-sonnet-5` → general quality
 

package/pipeline/skills/skills-index.md

@@ -126,7 +126,7 @@
 | core | `multi-agent-refactor` |  -  | Analyse the project: extract adapted best-practices, hunt real bugs + improvement areas, check upstream drift of derived skills, research th |
 | core | `multi-agent-resume` |  -  | Resume a stopped or failed task from the phase where it left off. Use when a task stopped or failed and should carry on from where it left o |
 | core | `multi-agent-resume-local` |  -  | Continue already-done LOCAL work through the pipeline tail: Review → Build+Test → Commit/PR → Report (technical analysis + Jira test-scenari |
-| core | `multi-agent-review` |  -  | Run parallel review on a branch diff or a Pull Request: 2 models on Claude Code (Fable + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonn |
+| core | `multi-agent-review` |  -  | Run parallel review on a branch diff or a Pull Request: 3 models on Claude Code (Fable + Opus + Sonnet), 3 models on Copilot CLI (GPT + Opus + Sonn |
 | core | `multi-agent-review-analysis` |  -  | Review a written analysis document instead of a diff: resolve it from a path, a Confluence page or a Jira issue, run the deterministic gates |
 | core | `multi-agent-review-issue` |  -  | Assess whether a GitHub issue is ready for multi-agent development: fetch it, grade scope / acceptance criteria / repro / design / API / sta |
 | core | `multi-agent-review-jira` |  -  | Assess whether a Jira issue is ready for multi-agent development: fetch it, grade scope / acceptance criteria / repro / design / API / stack |