opencode-swarm 7.114.0 → 7.114.1

package/.opencode/skills/writing-tests/SKILL.md

@@ -10,7 +10,7 @@ description: >
 
 # Writing Tests for opencode-swarm
 
-> **⚠️ Do NOT use the OpenCode `test_runner` tool to validate the full repo.** It is for targeted agent validation with explicit `files: [...]` or small targeted scopes. `scope: 'all'` requires `allow_full_suite: true` and is intended for opt-in CI mirrors only. Broad scopes can stall or kill OpenCode before the `MAX_SAFE_TEST_FILES = 50` guard in `src/tools/test-runner.ts` fires. For repo validation, use the shell commands in this file — per-file isolation loops match CI behavior. `allow_full_suite` should be used only when intentional and justified in the PR description. See [`AGENTS.md`](../../../AGENTS.md) invariant 6 for the full contract.
+> **⚠️ Do NOT use the OpenCode `test_runner` tool to validate the full repo.** It is for targeted agent validation with explicit `files: [...]` or small targeted scopes. `scope: 'all'` is gated behind the `SWARM_ALLOW_FULL_SUITE=1` env var (intended for opt-in CI mirrors only; there is no `allow_full_suite` arg). Broad scopes can stall or kill OpenCode before the `MAX_SAFE_TEST_FILES = 50` guard in `src/tools/test-runner.ts` fires. For repo validation, use the shell commands in this file — per-file isolation loops match CI behavior. See [`AGENTS.md`](../../../AGENTS.md) invariant 6 for the full contract.
 
 ## ⛔ STOP — Read Before Running Any Tests
 
@@ -522,22 +522,24 @@ const actual = readFileSync(path, 'utf-8').replace(/\r\n/g, '\n');
 
 ## CI Pipeline Structure
 
-The CI runs on three platforms (ubuntu, macos, windows). Tests are split into sequential steps within each platform's job.
+The CI runs on three platforms (ubuntu, macos, windows). Tests are split into 6 logical steps within each platform's job. (CI distributes files across shards via round-robin — see TESTING.md's CI Pipeline Steps table for the authoritative directory lists.)
 
 ```text
-Step 1: hooks — per-file isolation loop on every platform
-Step 2: cli — batch
-Step 3: commands + config — batch
-Step 4: tools — per-file isolation loop
-Step 5: services + build + quality + sast + sbom + scripts — per-file isolation loop
-Step 6: state + agents + knowledge + evidence + plan + misc — per-file isolation loop
+Step 1a: hooks (mock.module files — 15 files) — per-file isolation (dedicated step)
+Step 1b: hooks (remaining groups)               — per-file loop per group
+Step 2:  cli                                     — batch
+Step 3:  commands, config                        — batch
+Step 4:  tools                                   — per-file loop
+Step 5:  services, build, quality, sast, sbom, scripts — per-file loop
+Step 6:  adversarial, agents, background, context, diff, evidence, git, helpers,
+         knowledge, lang, output, parallel, plan, session, skills, types, utils — per-file loop
 ```
 
-**Steps 1 and 4-6 use per-file isolation:** each `.test.ts` file runs in its own `bun --smol` process to prevent `mock.module()` cache poisoning (#330). Steps 2-3 run files in batch (one process per step) because they have fewer mock conflicts.
+**Per-file isolation (steps 1a, 1b, 4-6):** each `.test.ts` file runs in its own `bun --smol` process to prevent `mock.module()` cache poisoning (#330). Steps 2-3 run files in batch because they have fewer mock conflicts. CI partitions the gated test set into **6 shards** round-robin per platform (no hardcoded file lists), with a per-file **retry budget** (two retries / three attempts before a failure is treated as real) and **quarantine** filters (`scripts/ci/quarantined-tests.txt`, plus `-macos`/`-windows` overrides) that drop known pre-existing failures.
 
 When writing a test, know which step your file will run in. In batch steps, do not assume isolation from other files in the same step.
 
-**Job timeout: 15 minutes.** A single hanging test will kill the entire platform's test run.
+**Job timeout: 40 minutes.** A hanging shard will kill the entire platform's test run; the per-file `run-test-with-timeout.ts` wrapper caps each file at 120 s (180 s kill-timeout).
 
 ## File Placement
 
@@ -695,6 +697,9 @@ When CI reports a `unit (ubuntu|macos|windows)` failure:
 - **Do not test framework behavior.** "Zod schema parses valid input" tests Zod, not your schema.
 - **Do not test test utilities.** If it only exists to support other tests, it doesn't need its own test.
 - **Do not mock everything.** If every dependency is mocked, you're testing the mock setup. Prefer real dependencies for pure functions and only mock I/O boundaries (filesystem, network, timers).
+- **Do not hardcode version numbers.** Version bumps are automated — a test asserting `version === '6.31.3'` breaks on every release.
+- **Do not use `sleep` or `setTimeout` for synchronization.** Use explicit signals, resolved promises, or `Bun.sleep()` with tight bounds.
+- **Do not spawn `cat /dev/zero`, `yes`, or other infinite-output commands.** Use `sleep 30` for "blocking command" tests.
 
 ### Anchored Content Assertions
 
@@ -718,9 +723,6 @@ Use this pattern for:
 - Critic outcome mappings in skill files (DROP, ASK_USER, RESOLVE, REPHRASE)
 - Classification category lists (self_resolved, user_decision, etc.)
 - Any structured section where word presence is necessary but position-dependent
-- **Do not hardcode version numbers.** Version bumps are automated — a test asserting `version === '6.31.3'` breaks on every release.
-- **Do not use `sleep` or `setTimeout` for synchronization.** Use explicit signals, resolved promises, or `Bun.sleep()` with tight bounds.
-- **Do not spawn `cat /dev/zero`, `yes`, or other infinite-output commands.** Use `sleep 30` for "blocking command" tests.
 
 ## Documented-Example Regression Tests
 
@@ -763,6 +765,84 @@ if (isWindows) test.skip('reason', () => {});
 - Use `.cmd` extension on Windows for npm/bun binaries: `process.platform === 'win32' ? 'bun.cmd' : 'bun'`.
 - Use array-form `spawn`/`spawnSync`, never shell string commands.
 
+### macOS rename-visibility race (write-then-read atomic files)
+
+On macOS/APFS, `fs.renameSync` can complete before the data is visible to
+subsequent reads. Tests that write-then-read atomic files may fail on
+`macos-latest` but pass on `ubuntu-latest` and `windows-latest`.
+
+**Symptom:** Test fails only on macOS with `result === null` or
+`result === undefined` immediately after a write that should have made the
+file visible.
+
+**Root cause:** macOS filesystem updates the directory entry asynchronously
+after `fs.renameSync`. Immediately-following reads may see ENOENT or stale
+data.
+
+**Fix — three layers (use all three for production code):**
+
+**Layer 1: Use `bunWrite` for atomic writes.** The `bunWrite` function in
+`src/utils/bun-compat.ts` already handles temp file creation, fsync,
+atomic rename, and parent directory fsync correctly across platforms.
+Do NOT reimplement this pattern.
+
+**Layer 2: Add ENOENT retry in the read path.** Wrap `validateSwarmPath` and
+the file read in a try/catch with a bounded retry loop for ENOENT:
+
+```typescript
+// CORRECT — retry on ENOENT only (not other errors), bounded
+const maxAttempts = 5;
+const retryDelayMs = 10;
+for (let attempt = 0; attempt < maxAttempts; attempt++) {
+  try {
+    const resolvedPath = _internals.validateSwarmPath(directory, filename);
+    const file = bunFile(resolvedPath);
+    const content = await file.text();
+    return content;
+  } catch (err) {
+    const isNotFound = (err as NodeJS.ErrnoException)?.code === 'ENOENT';
+    if (!isNotFound || attempt === maxAttempts - 1) {
+      return null;
+    }
+    await new Promise((resolve) => setTimeout(resolve, retryDelayMs));
+  }
+}
+return null;
+```
+
+CRITICAL: `validateSwarmPath` must be INSIDE the try block so that throws
+(for path traversal attempts) are caught and return null. Security tests
+expect `readSwarmFileAsync` to return null for traversal attempts.
+
+**Layer 3: Don't add arbitrary delays.** `setTimeout` should be bounded
+(5-10ms, max 5 attempts). Do not add `await new Promise(r => setTimeout(r, 100))`
+"just in case" — that's a code smell. The retry loop handles it.
+
+See [`.opencode/skills/engineering-conventions/SKILL.md`](../engineering-conventions/SKILL.md)
+for the evidence file flow that triggers this retry pattern in QA gates.
+
+### Node FileHandle API
+
+Node's `FileHandle` uses `.sync()`, NOT `.fsync()`:
+
+```typescript
+// CORRECT
+const fd = await fsPromises.open(dir, 'r');
+try {
+  await fd.sync();
+} finally {
+  await fd.close();
+}
+
+// WRONG — TypeScript error: Property 'fsync' does not exist on type 'FileHandle'
+const fd = await fsPromises.open(dir, 'r');
+try {
+  await fd.fsync();
+} finally {
+  await fd.close();
+}
+```
+
 ## Running Tests
 
 ### bash (Linux / macOS)
@@ -777,7 +857,7 @@ bun --smol test tests/unit/hooks --timeout 30000
 # Per-file loop (required for tools/services/agents — prevents mock poisoning)
 for f in tests/unit/tools/*.test.ts; do bun --smol test "$f" --timeout 30000; done
 
-# CI-equivalent run for batch steps
+# CI-equivalent run for batch steps (step 2: cli; step 3: commands + config)
 bun --smol test tests/unit/cli --timeout 120000
 bun --smol test tests/unit/commands tests/unit/config --timeout 120000
 ```
@@ -794,7 +874,7 @@ bun --smol test tests/unit/hooks --timeout 30000
 # Per-file loop (required for tools/services/agents — prevents mock poisoning)
 Get-ChildItem tests/unit/tools/*.test.ts | ForEach-Object { bun --smol test $_.FullName --timeout 30000 }
 
-# CI-equivalent run for batch steps
+# CI-equivalent run for batch steps (step 2: cli; step 3: commands + config)
 bun --smol test tests/unit/cli --timeout 120000
 bun --smol test tests/unit/commands tests/unit/config --timeout 120000
 
@@ -804,7 +884,7 @@ bun --smol test tests/unit/agents --timeout 60000 | Out-File "$env:TEMP\test_out
 
 **Note:** `for f in ...; do` bash syntax is invalid in PowerShell. Use `Get-ChildItem | ForEach-Object` instead. `Select-String -Last N` is also invalid — use `Select-Object -Last N`.
 
-**Warning:** Running `bun --smol test tests/unit/tools` as a single batch will cause mock poisoning failures. Always use the per-file loop for directories in CI steps 4-6 (tools, services, agents, etc.).
+**Warning:** Running `bun --smol test tests/unit/tools` as a single batch will cause mock poisoning failures. Always use the per-file loop for the per-file-isolation steps (1a, 1b, 4-6: tools, services, agents, etc.).
 
 The `--smol` flag reduces Bun's memory footprint. Use it when running large directories (50+ files).
 
@@ -821,17 +901,7 @@ The `--timeout 120000` flag sets per-test timeout to 120 seconds. Individual tes
 
 ## Known Pre-existing Test Failures
 
-The following test failures are pre-existing and unrelated to mock isolation:
-
-| Test file | Failures | Cause | Status |
-|-----------|----------|-------|--------|
-| `tests/unit/hooks/full-auto-intercept.test.ts` | 21/37 | `logger.log` returns early without `OPENCODE_SWARM_DEBUG=1` | Pre-existing |
-| `tests/unit/hooks/full-auto-intercept.dispatch.test.ts` | 2/46 | Same logger issue | Pre-existing |
-| `tests/unit/commands/help-compound-commands.test.ts` | Multiple | Command routing issues | Pre-existing |
-| `tests/unit/commands/index.test.ts` | Multiple | Command routing issues | Pre-existing |
-| `tests/unit/commands/issue-command.test.ts` | Multiple | Command routing issues | Pre-existing |
-| `src/__tests__/preflight-phase.test.ts` | 3/3 | `loadPlan` called twice per invocation (lines 930 + 545) | Bug exposed by cleanup |
-| `tests/unit/agents/architect-sounding-board-protocol.adversarial.test.ts` | 1 | Token budget threshold `35000` exceeded by prompt growth; soft regression indicator that prompt size needs attention | Pre-existing |
+Pre-existing and flaky failures are tracked in the per-platform quarantine ledgers (`scripts/ci/quarantined-tests.txt`, `quarantined-tests-macos.txt`, `quarantined-tests-windows.txt`), not in this skill. To confirm a failure is pre-existing, reproduce it in a clean worktree on `origin/main` (see the worktree verify protocol in `running-tests`).
 
 ## Known Cross-module mock.module Locations
 

package/README.md

@@ -1109,8 +1109,8 @@ Control how tool outputs are summarized for LLM context.
 | `/swarm analyze` | Analyze spec.md vs plan.md for requirement coverage gaps |
 | `/swarm sdd ...` | Inspect, validate, or project OpenSpec-compatible SDD artifacts into `.swarm/spec.md` |
 | `/swarm brainstorm [topic]` | Enter BRAINSTORM mode for structured requirement discovery before a spec |
-| `/swarm council <question> [--preset <name>] [--spec-review]` | Convene a multi-model General Council for advisory deliberation |
-| `/swarm issue <issue-url\|owner/repo#N\|N> [--plan] [--trace]` | Ingest a GitHub issue for localization and resolution |
+| `/swarm council <question> [--spec-review]` | Convene a multi-model General Council for advisory deliberation |
+| `/swarm issue <issue-url\|owner/repo#N\|N> [--plan] [--trace] [--no-repro]` | Ingest a GitHub issue for localization and resolution |
 | `/swarm pr-review <pr-url\|owner/repo#N\|N> [--council] [instructions...]` | Structured deep PR review with parallel lanes, reviewer confirmation, and critic challenge |
 | `/swarm pr-feedback [<pr-url\|owner/repo#N\|N>] [instructions...]` | Ingest and close known PR feedback (review comments, CI failures, conflicts) without a fresh review |
 | `/swarm pr subscribe <pr-url\|owner/repo#N\|N>` | Subscribe current session to PR monitoring (session-scoped); requires `pr_monitor.enabled: true` |

package/dist/cli/{curator-llm-factory-awp6xgvt.js → curator-llm-factory-26pyy8cq.js}

@@ -1,7 +1,7 @@
 // @bun
 import {
   createCuratorLLMDelegate
-} from "./index-zezjy05g.js";
+} from "./index-8phvk4v4.js";
 import"./index-g6va6n3t.js";
 import"./index-c8s9a3zh.js";
 import"./index-2148358e.js";

package/dist/cli/{curator-wvv628sv.js → curator-wwyj379g.js}

@@ -14,7 +14,7 @@ import {
   runCuratorInit,
   runCuratorPhase,
   writeCuratorSummary
-} from "./index-zezjy05g.js";
+} from "./index-8phvk4v4.js";
 import"./index-g6va6n3t.js";
 import"./index-c8s9a3zh.js";
 import"./index-2148358e.js";

package/dist/cli/{guardrail-explain-hhdtrm60.js → guardrail-explain-qzg7rr7f.js}

@@ -1,8 +1,8 @@
 // @bun
 import {
   handleGuardrailExplain
-} from "./index-jhdv0jdf.js";
-import"./index-zezjy05g.js";
+} from "./index-2aszzcyk.js";
+import"./index-8phvk4v4.js";
 import"./index-g6va6n3t.js";
 import"./index-c8s9a3zh.js";
 import"./index-2148358e.js";

package/dist/cli/{hive-promoter-1qkybw79.js → hive-promoter-kfjbydfj.js}

@@ -6,7 +6,7 @@ import {
   isHiveEligible,
   promoteFromSwarm,
   promoteToHive
-} from "./index-zezjy05g.js";
+} from "./index-8phvk4v4.js";
 import"./index-g6va6n3t.js";
 import"./index-c8s9a3zh.js";
 import"./index-2148358e.js";

package/dist/cli/{index-jhdv0jdf.js → index-2aszzcyk.js}

@@ -12,7 +12,7 @@ import {
   detectPosixWrites,
   detectWindowsWrites,
   resolveWriteTargets
-} from "./index-zezjy05g.js";
+} from "./index-8phvk4v4.js";
 import {
   checkFileAuthority,
   classifyFile,

package/dist/cli/{index-zezjy05g.js → index-8phvk4v4.js}

@@ -10432,11 +10432,11 @@ var _internals12 = {
     return KnowledgeConfigSchema2.parse({});
   },
   applyCuratorKnowledgeUpdates: async (directory, recommendations, knowledgeConfig) => {
-    const { applyCuratorKnowledgeUpdates: applyCuratorKnowledgeUpdates2 } = await import("./curator-wvv628sv.js");
+    const { applyCuratorKnowledgeUpdates: applyCuratorKnowledgeUpdates2 } = await import("./curator-wwyj379g.js");
     return applyCuratorKnowledgeUpdates2(directory, recommendations, knowledgeConfig);
   },
   checkHivePromotions: async (entries, knowledgeConfig) => {
-    const { checkHivePromotions } = await import("./hive-promoter-1qkybw79.js");
+    const { checkHivePromotions } = await import("./hive-promoter-kfjbydfj.js");
     return checkHivePromotions(entries, knowledgeConfig);
   },
   applyProposalTriage: async (directory, triage) => {
@@ -16774,10 +16774,9 @@ function parseArgs2(args) {
   return out;
 }
 var USAGE2 = [
-  "Usage: /swarm council <question> [--preset <name>] [--spec-review]",
+  "Usage: /swarm council <question> [--spec-review]",
   "",
   "  question         The question to put to the council",
-  "  --preset <name>  Use a named member preset from council.general.presets",
   "  --spec-review    Use spec_review mode (single advisory pass on a draft spec)",
   "",
   "Requires council.general.enabled: true and a configured search API key in the resolved config: global ~/.config/opencode/opencode-swarm.json, then project .opencode/opencode-swarm.json overrides."
@@ -17586,8 +17585,8 @@ var _internals24 = {
   loadCuratorDeps: async () => {
     const [{ CuratorConfigSchema }, curator, { createCuratorLLMDelegate: createCuratorLLMDelegate2 }] = await Promise.all([
       import("./schema-e5kd993s.js"),
-      import("./curator-wvv628sv.js"),
-      import("./curator-llm-factory-awp6xgvt.js")
+      import("./curator-wwyj379g.js"),
+      import("./curator-llm-factory-26pyy8cq.js")
     ]);
     return { CuratorConfigSchema, curator, createCuratorLLMDelegate: createCuratorLLMDelegate2 };
   }
@@ -18083,7 +18082,7 @@ import { fileURLToPath } from "url";
 // package.json
 var package_default = {
   name: "opencode-swarm",
-  version: "7.114.0",
+  version: "7.114.1",
   description: "Architect-centric agentic swarm plugin for OpenCode - hub-and-spoke orchestration with SME consultation, code generation, and QA review",
   main: "dist/index.js",
   types: "dist/index.d.ts",
@@ -18157,8 +18156,6 @@ var package_default = {
     ".opencode/skills/gate-attribution",
     ".opencode/skills/merge-queue-readiness",
     ".opencode/skills/skill-edit-validation",
-    ".opencode/skills/swarm",
-    ".opencode/skills/swarm-ci-monitor",
     ".opencode/skills/worktree-retry-cleanup",
     ".opencode/skills/test-file-split",
     ".opencode/skills/fork-pr-operations",
@@ -21785,7 +21782,7 @@ var USAGE6 = [
   "",
   "Flags:",
   "  --plan        Transition to plan creation after spec generation",
-  "  --trace       Run full fix-and-PR workflow (implies --plan)",
+  "  --trace       Run the fix workflow end-to-end (implies --plan); compose commit-pr to publish.",
   "  --no-repro    Skip reproduction step"
 ].join(`
 `);
@@ -28449,7 +28446,6 @@ var test_runner = createSwarmTool({
     coverage: exports_external.boolean().optional().describe("Enable coverage reporting if supported"),
     bail: exports_external.boolean().optional().describe("Stop running tests after the first failure. Default false. Note: coverage may be incomplete when bail=true with coverage=true."),
     timeout_ms: exports_external.number().optional().describe("Timeout in milliseconds (default 60000, max 300000)"),
-    allow_full_suite: exports_external.boolean().optional().describe('Explicit opt-in for scope "all". Required because full-suite output can destabilize SSE streaming.'),
     working_directory: exports_external.string().optional().describe("Explicit project root directory. When provided, tests run relative to this path instead of the plugin context directory. Use this when CWD differs from the actual project root.")
   },
   async execute(args, directory) {
@@ -32972,7 +32968,7 @@ function buildDetailedHelp(commandName, entry) {
 async function handleHelpCommand(ctx) {
   const targetCommand = ctx.args.join(" ");
   if (!targetCommand) {
-    const { buildHelpText } = await import("./index-p35ayncy.js");
+    const { buildHelpText } = await import("./index-gge9vh5s.js");
     return buildHelpText();
   }
   const tokens = targetCommand.split(/\s+/);
@@ -32981,7 +32977,7 @@ async function handleHelpCommand(ctx) {
     return _internals49.buildDetailedHelp(resolved.key, resolved.entry);
   }
   const similar = _internals49.findSimilarCommands(targetCommand);
-  const { buildHelpText: fullHelp } = await import("./index-p35ayncy.js");
+  const { buildHelpText: fullHelp } = await import("./index-gge9vh5s.js");
   if (similar.length > 0) {
     return `Command '/swarm ${targetCommand}' not found.
 
@@ -33114,7 +33110,7 @@ var COMMAND_REGISTRY = {
   },
   "guardrail explain": {
     handler: async (ctx) => {
-      const { handleGuardrailExplain } = await import("./guardrail-explain-hhdtrm60.js");
+      const { handleGuardrailExplain } = await import("./guardrail-explain-qzg7rr7f.js");
       return handleGuardrailExplain(ctx.directory, ctx.args);
     },
     description: "Dry-run: show what the guardrails would do to a command or write target (executes nothing)",
@@ -33124,7 +33120,7 @@ var COMMAND_REGISTRY = {
   },
   "guardrail-explain": {
     handler: async (ctx) => {
-      const { handleGuardrailExplain } = await import("./guardrail-explain-hhdtrm60.js");
+      const { handleGuardrailExplain } = await import("./guardrail-explain-qzg7rr7f.js");
       return handleGuardrailExplain(ctx.directory, ctx.args);
     },
     description: "Dry-run: show what the guardrails would do to a command or write target (executes nothing)",
@@ -33462,9 +33458,9 @@ Subcommands:
   },
   council: {
     handler: (ctx) => handleModeCommandWithBundledSkills(ctx, handleCouncilCommand),
-    description: "Enter architect MODE: COUNCIL \u2014 multi-model deliberation [question] [--preset <name>] [--spec-review]",
-    args: "<question> [--preset <name>] [--spec-review]",
-    details: "Triggers the architect to convene a three-agent General Council: Generalist (reviewer model), Skeptic (critic model), and Domain Expert (SME model). Use --preset <name> to choose a named member preset from council.general.presets. " + "The architect first runs 1\u20133 targeted web searches and passes a compiled RESEARCH CONTEXT " + "to all three agents before dispatching them in parallel. Agents deliberate using the NSED peer-review protocol (Round 1 independent analysis, Round 2 MAINTAIN/CONCEDE/NUANCE for disagreements). The architect synthesizes the final answer directly from convene_general_council output. --spec-review switches to single-pass advisory mode for spec review. Requires council.general.enabled: true and a search API key in the resolved config: global ~/.config/opencode/opencode-swarm.json, then project .opencode/opencode-swarm.json overrides.",
+    description: "Enter architect MODE: COUNCIL \u2014 multi-model deliberation [question] [--spec-review]",
+    args: "<question> [--spec-review]",
+    details: "Triggers the architect to convene a three-agent General Council: Generalist (reviewer model), Skeptic (critic model), and Domain Expert (SME model). " + "The architect first runs 1\u20133 targeted web searches and passes a compiled RESEARCH CONTEXT " + "to all three agents before dispatching them in parallel. Agents deliberate using the NSED peer-review protocol (Round 1 independent analysis, Round 2 MAINTAIN/CONCEDE/NUANCE for disagreements). The architect synthesizes the final answer directly from convene_general_council output. --spec-review switches to single-pass advisory mode for spec review. Requires council.general.enabled: true and a search API key in the resolved config: global ~/.config/opencode/opencode-swarm.json, then project .opencode/opencode-swarm.json overrides.",
     category: "agent",
     toolPolicy: "none"
   },
@@ -33599,7 +33595,7 @@ Subcommands:
     handler: (ctx) => handleModeCommandWithBundledSkills(ctx, handleIssueCommand),
     description: "Ingest a GitHub issue into the swarm workflow [url] [--plan] [--trace] [--no-repro]",
     args: "<issue-url|owner/repo#N|N> [--plan] [--trace] [--no-repro]",
-    details: "Triggers the architect to enter MODE: ISSUE_INGEST \u2014 ingests a GitHub issue, restructures it into a normalized intake note, localizes root cause through hypothesis-driven tracing, and outputs a resolution spec. --plan transitions to plan creation after spec generation. --trace runs the full fix-and-PR workflow (implies --plan). --no-repro skips the reproduction step. Supports full GitHub URL, owner/repo#N shorthand, or bare issue number (resolves against origin remote).",
+    details: "Triggers the architect to enter MODE: ISSUE_INGEST \u2014 ingests a GitHub issue, restructures it into a normalized intake note, localizes root cause through hypothesis-driven tracing, and outputs a resolution spec. --plan transitions to plan creation after spec generation. --trace runs the fix workflow end-to-end (implies --plan); compose commit-pr to publish. --no-repro skips the reproduction step. Supports full GitHub URL, owner/repo#N shorthand, or bare issue number (resolves against origin remote).",
     category: "agent",
     toolPolicy: "none"
   },
@@ -34872,7 +34868,7 @@ HARD CONSTRAINTS (apply regardless of skill load success):
 ### MODE: ISSUE_INGEST
 Activates when the user invokes /swarm issue <url> or the architect receives an ISSUE_INGEST signal.
 
-Purpose: Ingest issue evidence, trace impact, and transition to planning or tracing.
+Purpose: Ingest issue evidence, trace impact, and transition to the full fix workflow.
 
 ACTION: Load skill ${bundledProjectSkillFileReference("issue-ingest")} immediately. Follow the protocol defined there.
 

package/dist/cli/{index-p35ayncy.js → index-gge9vh5s.js}

@@ -1,7 +1,7 @@
 // @bun
 import {
   handleGuardrailExplain
-} from "./index-jhdv0jdf.js";
+} from "./index-2aszzcyk.js";
 import {
   handleGuardrailLog
 } from "./index-xbe1wtme.js";
@@ -81,7 +81,7 @@ import {
   handleWriteRetroCommand,
   normalizeSwarmCommandInput,
   resolveCommand
-} from "./index-zezjy05g.js";
+} from "./index-8phvk4v4.js";
 import"./index-g6va6n3t.js";
 import"./index-c8s9a3zh.js";
 import"./index-2148358e.js";

package/dist/cli/index.js

@@ -7,7 +7,7 @@ import {
   getPluginLockFilePaths,
   package_default,
   resolveCommand
-} from "./index-zezjy05g.js";
+} from "./index-8phvk4v4.js";
 import"./index-g6va6n3t.js";
 import"./index-c8s9a3zh.js";
 import"./index-2148358e.js";

package/dist/commands/council.d.ts

@@ -6,7 +6,6 @@
  * Deliberation → Moderator Pass → Output).
  *
  * Flag parsing:
- *   --preset <name>   → emits "[MODE: COUNCIL preset=<name>] <question>"
  *   --spec-review     → emits "[MODE: COUNCIL spec_review] <question>"
  *   default           → emits "[MODE: COUNCIL] <question>"
  *   no args           → returns usage string (no throw)

package/dist/commands/registry.d.ts

@@ -490,8 +490,8 @@ export declare const COMMAND_REGISTRY: {
     };
     readonly council: {
         readonly handler: (ctx: CommandContext) => CommandResult;
-        readonly description: "Enter architect MODE: COUNCIL — multi-model deliberation [question] [--preset <name>] [--spec-review]";
-        readonly args: "<question> [--preset <name>] [--spec-review]";
+        readonly description: "Enter architect MODE: COUNCIL — multi-model deliberation [question] [--spec-review]";
+        readonly args: "<question> [--spec-review]";
         readonly details: string;
         readonly category: "agent";
         readonly toolPolicy: "none";
@@ -627,7 +627,7 @@ export declare const COMMAND_REGISTRY: {
         readonly handler: (ctx: CommandContext) => CommandResult;
         readonly description: "Ingest a GitHub issue into the swarm workflow [url] [--plan] [--trace] [--no-repro]";
         readonly args: "<issue-url|owner/repo#N|N> [--plan] [--trace] [--no-repro]";
-        readonly details: "Triggers the architect to enter MODE: ISSUE_INGEST — ingests a GitHub issue, restructures it into a normalized intake note, localizes root cause through hypothesis-driven tracing, and outputs a resolution spec. --plan transitions to plan creation after spec generation. --trace runs the full fix-and-PR workflow (implies --plan). --no-repro skips the reproduction step. Supports full GitHub URL, owner/repo#N shorthand, or bare issue number (resolves against origin remote).";
+        readonly details: "Triggers the architect to enter MODE: ISSUE_INGEST — ingests a GitHub issue, restructures it into a normalized intake note, localizes root cause through hypothesis-driven tracing, and outputs a resolution spec. --plan transitions to plan creation after spec generation. --trace runs the fix workflow end-to-end (implies --plan); compose commit-pr to publish. --no-repro skips the reproduction step. Supports full GitHub URL, owner/repo#N shorthand, or bare issue number (resolves against origin remote).";
         readonly category: "agent";
         readonly toolPolicy: "none";
     };