@haposoft/cafekit 0.16.7 → 0.18.0

package/README.md

@@ -2,7 +2,7 @@
 
 > Native spec-driven workflow and runtime bundle for Claude Code and Codex CLI.
 
-[![Version](https://img.shields.io/badge/version-0.16.7-blue.svg)](https://github.com/haposoft/cafekit)
+[![Version](https://img.shields.io/badge/version-0.18.0-blue.svg)](https://github.com/haposoft/cafekit)
 [![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)
 [![Claude%20Code](https://img.shields.io/badge/Claude%20Code-Native-orange.svg)](https://claude.ai/code)
 [![Codex%20CLI](https://img.shields.io/badge/Codex%20CLI-Native-111111.svg)](https://developers.openai.com/codex)
@@ -147,6 +147,13 @@ cat .codex/cafekit.json
 The Claude statusline reads three keys from `.claude/runtime.json`:
 
 - `"statusline"`: render mode — `full` (default), `compact`, `minimal`, or `none`.
+  The default is one line, `◆ model (effort) ϟ✱  ◔ context%  ⧗ 5h quota% reset  ◷ weekly%  ⎇ branch ●changed ↑↓ #PR✓ ⊟  ± +added/-removed  ⏱ session  ↻ cache hit%`
+  (`ϟ` fast mode, `✱` thinking, `#PR` with ✓ approved / ✗ changes requested, `⊟` inside a git worktree;
+  each appears only when present), which drops cache, session time, changes, weekly, quota, then the
+  branch as the terminal narrows. Quota comes only from the `rate_limits` Claude Code sends (Pro/Max,
+  after a session's first API response); without it the quota segments are hidden. `full` adds
+  agent and todo lines only while they exist, `compact` is the line alone, and `minimal`
+  keeps model, context and branch.
 - `statuslineColors`: set `false` to disable ANSI colors (`NO_COLOR` also wins).
 - `statuslineLayout`: optional custom layout. Global shape
   `{ "lines": [["model", "context"], ["directory", "git"]] }` — each inner
@@ -156,8 +163,8 @@ The Claude statusline reads three keys from `.claude/runtime.json`:
   ids are ignored; an empty or invalid layout falls back to the default
   renderers, and leaving the key out keeps the default output unchanged. The
   `cost` section renders only when enabled here, billing mode is API, and cost
-  data exists. Quota shows five-hour and weekly windows only while the usage
-  cache is fresh.
+  data exists. Quota shows five-hour and weekly windows only while Claude Code
+  sends `rate_limits`. The `usage` key is deprecated and has no effect.
 
 ## Core Skills
 
@@ -167,8 +174,6 @@ CafeKit ships many skills, but the main release surface is:
 - `/cf:scout <search-target>`: discover relevant files locally first, delegating only for permitted broad independent scopes
 - `/cf:brainstorm <idea-or-problem>`: scout the repo, clarify exact requirements, compare approaches, and hand off to specs
 - `/cf:research <decision>`: choose proportional depth and return traceable evidence for an uncertain technical decision
-- `/cf:route <material-request>`: classify ambiguous, multi-step, multi-domain, or elevated-risk work and compose the shortest valid installed chain
-- `/cf:loop <bounded-experiment>`: run explicit-only numeric optimization in an isolated worktree and return a base-bound patch handoff
 - `/cf:orca <orca-request>`: inside an Orca (onorca.dev) pane, route a request to read, wait on, or send to another agent pane, or spawn one into a worktree
 - `/cf:specs <feature-description>`: create or resume a structured spec workflow
 - `/cf:develop <feature-name>`: implement from approved spec artifacts
@@ -194,7 +199,7 @@ completes work. Chat is the default output. A durable file needs explicit user
 authority, and no Brainstorm output is live proof or Specs/Develop approval or
 execution authority.
 
-### Adaptive Research and bounded Loop
+### Adaptive Research
 
 Use `/cf:research` when the result is a decision: Quick, Standard, or Deep
 research binds material claims to a URL or repository anchor, authority,
@@ -202,18 +207,9 @@ date/version, applicability, and `confirmed`, `inferred`, or `unresolved`
 state. Research returns evidence and tradeoffs; it does not implement the
 recommendation or guarantee correctness.
 
-Use `/cf:loop` only by explicit request and only after its preflight freezes
-Goal, isolated Scope, finite numeric Metric and Direction, reproducible
-Baseline, distinct Guard, noise policy and minimum delta, budget, and stop
-conditions. Each iteration stays in a detached worktree. The result is a
-base-bound isolated patch handoff, never an automatic apply, commit, push, or
-guarantee of improvement.
-
 ```text
 /cf:research Compare current local-first search libraries for this repository
-/cf:loop Goal="reduce parser latency" Scope="packages/parser" Metric="median ms, lower" Baseline="pinned base" Guard="npm test" Noise="MAD; minimum delta 2%" Budget="10 iterations" Stop="budget, drift, or failed guard"
 $cf-research Compare current local-first search libraries for this repository
-$cf-loop Goal="reduce parser latency" Scope="packages/parser" Metric="median ms, lower" Baseline="pinned base" Guard="npm test" Noise="MAD; minimum delta 2%" Budget="10 iterations" Stop="budget, drift, or failed guard"
 ```
 
 Optional document skills are `docs`, `docx`, `pdf`, `pptx`, `xlsx`, and
@@ -244,7 +240,7 @@ node .claude/scripts/generate-skill-catalog.cjs --skills
 Routing uses progressive disclosure: live catalog metadata selects a candidate,
 the selected `SKILL.md` supplies its contract, and only relevant references are
 loaded. Invoke an installed skill directly for explicit or obvious low-risk
-work; use `cf:route` only when classification or chaining is material. When a
+work; compose a short chain only when classification or chaining is material. When a
 capability or agent is absent, continue inline when safe or name the gap.
 Hooks are not a skill router and never grant capability presence or authority. Source
 and installed projections are verified; live-model adherence remains
@@ -310,17 +306,13 @@ specs/<feature-name>/
 ```
 
 The process-first workflow has three human decision gates. `cf:specs` opens
-C1 to fix scope before authoring and C2 to resolve adversarial findings, then
+GATE-SCOPE to fix scope before authoring and GATE-REVIEW to resolve adversarial findings, then
 stops. A later explicit `cf:develop` invocation executes the tasks and
-presents C3 for closeout only after real execution proof. Each task has one
+presents GATE-DONE for closeout only after real execution proof. Each task has one
 `Status:` field and keeps its canonical `## Receipt` inline, including the
 exact Verification Plan command, `Exit: 0`, `Verification: PASS`,
 runtime-derived Base and Head, and current command output.
 
-Existing packets with `spec.json`, nested `tasks/task-R*.md`, or separate
-receipts remain supported through the legacy compatibility adapter. New Specs
-work does not author those legacy artifacts.
-
 ## Development
 
 Run package self-tests:

package/bin/__tests__/codex-hooks-ownership.test.js

@@ -136,8 +136,8 @@ test('repeated installs never duplicate a CafeKit hook', () => {
     const config = readHooks(root);
     assert.equal(allHandlers(config).length, afterFirst, 'a reinstall added handlers that were already registered');
 
-    // Uniqueness is per event, not global: state.cjs and completion-authority.cjs each
-    // register under several events by design.
+    // Uniqueness is per event, not global: state.cjs registers under several events by
+    // design.
     for (const [eventName, groups] of Object.entries(config.hooks)) {
       const scripts = groups
         .flatMap((group) => group.hooks || [])

package/bin/__tests__/codex-hooks.test.js

@@ -611,24 +611,6 @@ test('Codex inspect hook scans native shell commands and structured paths', () =
   });
 });
 
-test('Codex task scaffold hook blocks apply_patch Add File with stderr guidance', () => {
-  inHookFixture((root, hooks) => {
-    const result = runHook(path.join(hooks, 'task-scaffold-guard.cjs'), root, {
-      cwd: root,
-      session_id: 'session-a',
-      hook_event_name: 'PreToolUse',
-      tool_name: 'apply_patch',
-      tool_input: {
-        command: '*** Begin Patch\n*** Add File: specs/auth/tasks/task-R0-01-auth.md\n+x\n*** End Patch\n'
-      }
-    });
-
-    assert.equal(result.status, 2);
-    assert.match(result.stderr, /TASK SCAFFOLD REQUIRED/);
-    assert.match(result.stderr, /node \.codex\/scripts\/spec-scaffold\.cjs auth/);
-  });
-});
-
 test('Codex completion gate cannot be bypassed from a nested cwd', () => {
   inHookFixture((root, hooks) => {
     const nested = path.join(root, 'packages', 'app');
@@ -1036,24 +1018,6 @@ test('Codex state hook persists direct native event fields at project root', ()
   });
 });
 
-test('semantic review authority hook is registered only for SubagentStop on Claude and Codex', () => {
-  const configurations = [
-    JSON.parse(fs.readFileSync(path.join(PACKAGE_ROOT, 'src/claude/settings/settings.json'), 'utf8')).hooks,
-    JSON.parse(fs.readFileSync(path.join(PACKAGE_ROOT, 'src/codex/hooks.json'), 'utf8')).hooks,
-  ];
-  for (const hooks of configurations) {
-    const registrations = [];
-    for (const [event, groups] of Object.entries(hooks)) {
-      for (const group of groups) {
-        for (const hook of group.hooks || []) {
-          if (String(hook.command || '').includes('semantic-review-authority.cjs')) registrations.push(event);
-        }
-      }
-    }
-    assert.deepEqual(registrations, ['SubagentStop']);
-  }
-});
-
 test('Codex canonical receipt provenance requires both Base and Head', () => {
   const receipt = require(path.join(CODEX_HOOKS, 'lib', 'spec-receipt.cjs'));
   const onlyBase = receipt.validateCanonicalReceipt('Verification: PASS\nCommand: pnpm test\nExit: 0\nBase: abc\n');

package/bin/__tests__/codex-native.test.js

@@ -69,7 +69,7 @@ const DOCS_ADAPTIVE_BUNDLE = [
   'references/update-workflow.md',
   'references/standard-docs-workflow.md'
 ];
-const HOTFIX_SOURCE_ROOT = path.join(PACKAGE_ROOT, 'src/claude/skills/hotfix');
+const HOTFIX_SOURCE_ROOT = path.join(PACKAGE_ROOT, 'src/claude/skills/fix');
 const HOTFIX_BUNDLE = [
   'SKILL.md',
   'references/diagnosis-protocol.md',
@@ -111,7 +111,7 @@ function boundaryTableHasRequiredRows(table) {
 }
 
 const IMPLEMENTATION_READINESS_CLAUSES = {
-  noInvention: 'Before implementation handoff, apply the **no-invention gate**: if two implementations conform to the packet text yet can produce different externally observable output, state, error, security, or compatibility behavior, surface the missing choice as an explicit C1 or C2 question and block handoff.',
+  noInvention: 'Before implementation handoff, apply the **no-invention gate**: if two implementations conform to the packet text yet can produce different externally observable output, state, error, security, or compatibility behavior, surface the missing choice as an explicit GATE-SCOPE or GATE-REVIEW question and block handoff.',
   materialDefinition: 'A boundary is material when the task creates, changes, or depends on it and a different choice changes an external observation, security, durable data, compatibility, or proof reachability. Require only the matching material row; omit nonmaterial categories.',
   exactBoundaryChoices: 'For every required row, name each listed choice exactly; labels such as “JSON”, “local path”, “locked”, or “timestamped” alone remain unresolved.',
   proofPlanLines: [
@@ -130,11 +130,11 @@ const IMPLEMENTATION_READINESS_CLAUSES = {
   disposableReviewControls: 'Run mutation or destructive negative controls only on disposable copies below a verified temporary root, never tracked worktree or canonical source bytes.',
   failureSemantics: '`Crash` means abrupt unhandled termination before the claimed catch point; a catchable failure returns/raises an error or exits nonzero. Never use them interchangeably.',
   privacyIdentifiers: 'Any privacy/security claim names the exact identifier surface at risk, such as an env var, header, path, token class, or field name; generic “sensitive data” is insufficient.',
-  freshReplay: 'After applying an accepted C2 finding, a fresh-context closure pass records and freshly replays its original counterexample after the repair under this exact review-log header:',
+  freshReplay: 'After applying an accepted GATE-REVIEW finding, a fresh-context closure pass records and freshly replays its original counterexample after the repair under this exact review-log header:',
   distinctRepairProof: '`Repaired at` cites the repair edit; `Proved at` must cite distinct evidence from the fresh replay, never the repair-edit citation.',
   closureTransition: 'An accepted finding transitions `accepted → repaired → PASS|FAIL|UNKNOWN`.',
   unknownBlocks: 'Only `PASS` closes it; `FAIL` remains open for the remaining paper-review round; `UNKNOWN` blocks implementation handoff.',
-  scopeReturnsToC1: 'A repair that adds user semantics or scope returns to C1.'
+  scopeReturnsToC1: 'A repair that adds user semantics or scope returns to GATE-SCOPE.'
 };
 const ADAPTIVE_COVERAGE_PROFILE_HEADER = [
   'ID', 'Outcome', 'Change kinds', 'Material surfaces', 'Ambiguity/action',
@@ -148,7 +148,7 @@ const ADAPTIVE_COVERAGE_AMBIGUITY_ROWS = [
   ['State', 'Required action'],
   ['`none`', 'proceed'],
   ['`examples-needed`', 'add two or three examples only for an already decided rule; promote to `decision-needed` if an example changes observable behavior'],
-  ['`decision-needed`', 'ask the user at C1/C2 and keep affected tasks blocked'],
+  ['`decision-needed`', 'ask the user at GATE-SCOPE/GATE-REVIEW and keep affected tasks blocked'],
   ['`design-needed`', 'after user-owned decisions settle, route material competing technical designs through Brainstorm']
 ];
 const ADAPTIVE_REVIEWER_ROWS = [
@@ -169,12 +169,12 @@ const ADAPTIVE_COVERAGE_CLAUSES = {
   openKinds: 'Change kinds are multi-valued (`add`, `modify`, `fix`, `refactor`, `remove`, `migrate`, `integrate`), and unfamiliar kinds or surfaces use `other:<verbatim>` rather than disappearing.',
   scopedUnion: 'Each task references its CP IDs; authoring, review, edge, and proof obligations union only inside affected rows/tasks.',
   profileRederivation: 'Rederive affected CP rows after any accepted scope, outcome, criteria, ownership, dependency, risk, or proof delta before task status.',
-  plannedProof: '`Required proof` is a planned level set, not execution\nevidence: known but unrun proof may be `pending`; `UNKNOWN` reachability blocks\n`pending`; missing, failed, or unavailable required evidence blocks `done`/C3.',
+  plannedProof: '`Required proof` is a planned level set, not execution\nevidence: known but unrun proof may be `pending`; `UNKNOWN` reachability blocks\n`pending`; missing, failed, or unavailable required evidence blocks `done`/GATE-DONE.',
   proofSeparation: 'Levels stay separate and never promote one another.',
   liveLimit: 'Source/static checks prove the written contract, not live-model adherence.',
   specMakerAuthority: 'they are the canonical risk and coverage authority. Do not duplicate\ntheir taxonomy here.',
   specMakerAmbiguity: 'Apply the canonical ambiguity action; examples never decide observable behavior.',
-  specMakerRoute: 'Apply their risk-first route before C1 and stop when the\nrequest qualifies for direct work; hand off when it requires Brainstorm-only exploration.',
+  specMakerRoute: 'Apply their risk-first route before GATE-SCOPE and stop when the\nrequest qualifies for direct work; hand off when it requires Brainstorm-only exploration.',
   reviewRisk: 'Keep Fact Checker as the baseline. Assign every remaining material CP risk to a\nnamed reviewer lens; a critical row includes both relevant security-adversary and\nfailure-mode coverage, and nonmaterial lenses are not added.',
   reviewCapacity: 'Reviewer count is fixed by the table, not lens count. Give each reviewer a distinct primary lens; when material lenses exceed reviewers, combine related named lenses on one reviewer and keep every material lens assigned.'
 };
@@ -558,7 +558,7 @@ function authoringProjectionIssues(files) {
     || !files.templates.includes('task-NN-<slug>.md')) {
     issues.push('flat-layout-template');
   }
-  for (const gate of ['C1 — Scope', 'C2 — Findings', 'C3 — Done']) {
+  for (const gate of ['GATE-SCOPE', 'GATE-REVIEW', 'GATE-DONE']) {
     if (!files.skill.includes(gate)) issues.push(`human-gate-${gate.slice(0, 2).toLowerCase()}`);
   }
   if (!files.review.includes('fresh-context red team')
@@ -596,7 +596,7 @@ function authoringProjectionIssues(files) {
   }
   if (!files.codex.includes('flat `task-NN-*.md` files')
     || !files.codex.includes('inline `## Receipt`')
-    || !files.codex.includes('C1') || !files.codex.includes('C2') || !files.codex.includes('C3')) {
+    || !files.codex.includes('GATE-SCOPE') || !files.codex.includes('GATE-REVIEW') || !files.codex.includes('GATE-DONE')) {
     issues.push('codex-process-v3');
   }
   return [...new Set(issues)].sort();
@@ -1203,8 +1203,8 @@ function assertInstalledAuthoringProjection(root, installedRoot) {
       name: 'fresh-original-counterexample-replay',
       issue: 'repair-closure',
       relative: 'references/review.md',
-      from: 'After applying an accepted C2 finding, a fresh-context closure pass records and\nfreshly replays its original counterexample after the repair under this exact review-log header:',
-      to: 'After applying an accepted C2 finding, record the repair under this review-log header:'
+      from: 'After applying an accepted GATE-REVIEW finding, a fresh-context closure pass records and\nfreshly replays its original counterexample after the repair under this exact review-log header:',
+      to: 'After applying an accepted GATE-REVIEW finding, record the repair under this review-log header:'
     },
     {
       name: 'distinct-repair-and-proof-evidence',
@@ -1313,7 +1313,7 @@ function assertInstalledAuthoringProjection(root, installedRoot) {
     ['stale-profile-after-c2', 'templates', ADAPTIVE_COVERAGE_CLAUSES.profileRederivation,
       'Keep existing coverage rows after accepted plan changes.', ['profile-lifecycle']],
     ['planned-proof-blocks-start', 'templates', ADAPTIVE_COVERAGE_CLAUSES.plannedProof,
-      '`Required proof` is execution evidence: known but unrun proof blocks `pending`; `UNKNOWN` may proceed; missing evidence may still reach `done`/C3.', ['proof-lifecycle']],
+      '`Required proof` is execution evidence: known but unrun proof blocks `pending`; `UNKNOWN` may proceed; missing evidence may still reach `done`/GATE-DONE.', ['proof-lifecycle']],
     ['source-promotes-live', 'templates', ADAPTIVE_COVERAGE_CLAUSES.proofSeparation,
       'Source proof may promote installed and live proof.', ['proof-lifecycle']],
     ['unmapped-proof-level', 'templates', IMPLEMENTATION_READINESS_CLAUSES.proofLevelMapping,
@@ -1327,7 +1327,7 @@ function assertInstalledAuthoringProjection(root, installedRoot) {
     ['spec-maker-examples-decide', 'specMaker', ADAPTIVE_COVERAGE_CLAUSES.specMakerAmbiguity,
       'Use examples to settle every ambiguous observable behavior.', ['spec-maker-authority']],
     ['spec-maker-skips-brainstorm', 'specMaker', ADAPTIVE_COVERAGE_CLAUSES.specMakerRoute,
-      'Apply the risk-first route before C1 and stop only for direct work.', ['spec-maker-authority']],
+      'Apply the risk-first route before GATE-SCOPE and stop only for direct work.', ['spec-maker-authority']],
     ['static-proves-live', 'skill', ADAPTIVE_COVERAGE_CLAUSES.liveLimit,
       'Source/static checks prove live-model adherence.', ['proof-lifecycle']]
   ];
@@ -1561,13 +1561,15 @@ test('Codex structured-input corpus oracle stays differential and production-awa
     'src/claude/agents/spec-maker.md',
     'src/claude/hooks/session.cjs',
     'src/claude/skills/git/SKILL.md',
-    'src/claude/skills/inspect/SKILL.md'
+    'src/claude/skills/scout/SKILL.md',
+    'src/claude/skills/sync/references/rebind-and-audit.md'
   ];
   assert.deepEqual(actualPaths, expectedPaths, 'every source occurrence needs an explicit projection oracle');
 
   const instructionExpectedSnippets = new Map([
     ['src/claude/skills/git/SKILL.md', 'Present options via a structured user-input request — header'],
-    ['src/claude/skills/inspect/SKILL.md', '**Fallback to a structured user-input request:**']
+    ['src/claude/skills/scout/SKILL.md', '**Fallback to a structured user-input request:**'],
+    ['src/claude/skills/sync/references/rebind-and-audit.md', 'End by asking with a structured user-input request when the host has']
   ]);
   for (const [relativePath, expectedSnippet] of instructionExpectedSnippets) {
     const sourcePath = path.join(PACKAGE_ROOT, relativePath);
@@ -1983,8 +1985,7 @@ test('Codex Windows hook launchers stay project-bound without Git from nested cw
     );
     const launchers = allHookLaunchers(config);
     assert.ok(launchers.length > 0, 'installed Codex config must register hook launchers');
-    const semanticReviewEvents = [];
-    for (const { event, handler } of launchers) {
+    for (const { handler } of launchers) {
       assert.doesNotMatch(handler.commandWindows, /\$\(/);
       assert.doesNotMatch(handler.commandWindows, /\bgit\b/i);
       assert.doesNotMatch(handler.commandWindows, /process\.cwd\(\)|existsSync/);
@@ -1997,11 +1998,7 @@ test('Codex Windows hook launchers stay project-bound without Git from nested cw
         true,
         `missing installed hook: ${path.basename(target)}`
       );
-      if (path.basename(target) === 'semantic-review-authority.cjs') {
-        semanticReviewEvents.push(event);
-      }
     }
-    assert.deepEqual(semanticReviewEvents, ['SubagentStop']);
 
     const nested = path.join(projectRoot, 'nested', 'workspace');
     fs.mkdirSync(nested, { recursive: true });
@@ -2260,24 +2257,24 @@ test('Codex installed Specs and spec-maker reject adaptive coverage mutations',
     assert.ok(skillNames.every((name) => /^[a-z0-9-]+$/.test(name)));
 
     const scoutSkill = fs.readFileSync(
-      path.join(root, '.agents', 'skills', 'inspect', 'SKILL.md'),
+      path.join(root, '.agents', 'skills', 'scout', 'SKILL.md'),
       'utf8'
     );
     assert.match(scoutSkill, /^name:\s*cf-scout$/m);
     assert.equal(
-      fs.existsSync(path.join(root, '.agents', 'skills', 'scout')),
+      fs.existsSync(path.join(root, '.agents', 'skills', 'inspect')),
       false,
-      'public Scout rename must keep the manifest-owned inspect directory'
+      'Scout lives in the scout directory; the retired inspect directory must not be installed'
     );
     const askSkill = fs.readFileSync(
-      path.join(root, '.agents', 'skills', 'question', 'SKILL.md'),
+      path.join(root, '.agents', 'skills', 'ask', 'SKILL.md'),
       'utf8'
     );
     assert.match(askSkill, /^name:\s*cf-ask$/m);
     assert.equal(
-      fs.existsSync(path.join(root, '.agents', 'skills', 'ask')),
+      fs.existsSync(path.join(root, '.agents', 'skills', 'question')),
       false,
-      'public Ask rename must keep the manifest-owned question directory'
+      'Ask lives in the ask directory; the retired question directory must not be installed'
     );
 
     const catalog = spawnSync(
@@ -2338,7 +2335,7 @@ test('Codex installed Specs and spec-maker reject adaptive coverage mutations',
       (key.startsWith('.codex/') || key.startsWith('.agents/')) && !key.includes('../')
     )));
 
-    const questionSkill = path.join(root, '.agents', 'skills', 'question', 'SKILL.md');
+    const questionSkill = path.join(root, '.agents', 'skills', 'ask', 'SKILL.md');
     fs.appendFileSync(questionSkill, '\nUSER-CODEX-SENTINEL\n');
     // Fresh installs are exact. Refresh/upgrade intentionally documents the
     // current Codex limitation: removed skill paths are not pruned.
@@ -2426,7 +2423,7 @@ test('Codex installed Brainstorm skill reference and agent preserve proportional
     const mutations = [
       {
         name: 'adaptive-direct-precedence-removed', source: 'skill',
-        from: 'Route Direct first, then\napply controls only to requests that remain in Brainstorm.',
+        from: 'Route Direct first, then apply controls only to requests that remain in Brainstorm.',
         to: 'Apply controls before Direct classification.', expected: ['adaptive-direct-precedence']
       },
       {
@@ -2441,37 +2438,37 @@ test('Codex installed Brainstorm skill reference and agent preserve proportional
       },
       {
         name: 'adaptive-failure-isolation-removed', source: 'skill',
-        from: 'failure isolation for partial or cascading failure\nacross boundaries',
+        from: 'failure isolation for partial or cascading failure across boundaries',
         to: 'generic failure notes', expected: ['adaptive-lens-trigger-skip']
       },
       {
         name: 'adaptive-evidence-fallback-removed', source: 'skill',
-        from: 'Missing evidence\nforces feasibility `unknown` and confidence `low`.',
+        from: 'Missing evidence forces feasibility `unknown` and confidence `low`.',
         to: 'Missing evidence permits a confident guess.', expected: ['adaptive-evidence-semantics']
       },
       {
         name: 'adaptive-numeric-evidence-removed', source: 'skill',
-        from: 'A numeric estimate requires\nrange, unit, basis, evidence, and assumptions; otherwise report `unknown`.',
+        from: 'A numeric estimate requires range, unit, basis, evidence, and assumptions; otherwise report `unknown`.',
         to: 'A numeric estimate may be a best-effort number.', expected: ['adaptive-numeric-estimates']
       },
       {
         name: 'adaptive-pre-tool-redaction-removed', source: 'skill',
-        from: 'Before an\nexternal visual tool or adviser handoff, minimize context and redact secrets,\ncredentials, private keys, access tokens, and unnecessary PII.',
+        from: 'Before an external visual tool or adviser handoff, minimize context and redact secrets, credentials, private keys, access tokens, and unnecessary PII.',
         to: 'Forward full context to every external tool and adviser.', expected: ['adaptive-pre-tool-redaction']
       },
       {
         name: 'adaptive-adviser-gate-removed', source: 'skill',
-        from: '`--advice` invokes\n`brainstormer` only after the material-choice gate;',
+        from: '`--advice` invokes `brainstormer` only after the material-choice gate;',
         to: '`--advice` invokes `brainstormer` before routing;', expected: ['adaptive-adviser-gate']
       },
       {
         name: 'adaptive-decision-freshness-removed', source: 'skill',
-        from: 'The first section records target\nidentity, current source revision and worktree state or `[UNVERIFIED]`, an\nevidence-as-of value, and what change invalidates the brief.',
+        from: 'The first section records target identity, current source revision and worktree state or `[UNVERIFIED]`, an evidence-as-of value, and what change invalidates the brief.',
         to: 'The handoff has no revision or freshness binding.', expected: ['adaptive-decision-freshness']
       },
       {
         name: 'adaptive-overlay-authority-removed', source: 'skill',
-        from: 'Neither overlay\nwrites, approves, persists, dispatches, or completes work.',
+        from: 'Neither overlay writes, approves, persists, dispatches, or completes work.',
         to: 'Overlays may persist, approve, dispatch, and complete work.', expected: ['adaptive-non-authority']
       },
       {
@@ -2491,7 +2488,7 @@ test('Codex installed Brainstorm skill reference and agent preserve proportional
       {
         name: 'exploration-stop-removed',
         source: 'skill',
-        from: 'Do not request design approval, persist a report, or\n   invoke another workflow without a new explicit request.',
+        from: 'Do not request design approval, persist a report, or invoke another workflow without a new explicit request.',
         to: 'Request approval, persist a report, and invoke another workflow.',
         expected: ['exploration-stop']
       },
@@ -2841,7 +2838,7 @@ function hotfixProjectionIssues(files) {
   }
   if (!parallel.includes('Diagnosis still starts only')
     || !parallel.includes('after the required scout outputs are synthesized')
-    || !parallel.includes('Research begins only after Step 2 diagnosis')
+    || !parallel.includes('Research begins only after diagnosis')
     || parallel.includes('scout + diagnose + research together')
     || parallel.includes("You don't need to wait for scouting")) {
     issues.add('scout-before-diagnosis');
@@ -2858,9 +2855,9 @@ test('Codex installed Fix preserves the adaptive repair contract', () => {
     const result = install(root);
     assert.equal(result.status, 0, `${result.stdout}\n${result.stderr}`);
 
-    const installedRoot = path.join(root, '.agents/skills/hotfix');
-    assert.equal(fs.existsSync(path.join(root, '.agents/skills/fix')), false,
-      'public Fix rename must keep the manifest-owned hotfix directory');
+    const installedRoot = path.join(root, '.agents/skills/fix');
+    assert.equal(fs.existsSync(path.join(root, '.agents/skills/hotfix')), false,
+      'Fix lives in the fix directory; the retired hotfix directory must not be installed');
     for (const relative of HOTFIX_BUNDLE) {
       const sourcePath = path.join(HOTFIX_SOURCE_ROOT, relative);
       assert.equal(
@@ -2952,7 +2949,7 @@ test('Codex installed Fix preserves the adaptive repair contract', () => {
       },
       {
         name: 'research-starts-before-diagnosis', file: 'references/parallel-patterns.md',
-        from: 'Research begins only after Step 2 diagnosis', to: 'Research may begin before Step 2 diagnosis',
+        from: 'Research begins only after diagnosis', to: 'Research may begin before diagnosis',
         expected: ['scout-before-diagnosis']
       }
     ];
@@ -3162,16 +3159,14 @@ test('Codex install on top of existing Claude installation preserves content and
   });
 });
 
-test('Codex installed code_auditor contains Strict conditional marker', () => {
+test('Codex installed code_auditor carries no Strict attestation marker', () => {
   inTempProject((root) => {
     const result = install(root);
     assert.equal(result.status, 0, `${result.stdout}\n${result.stderr}`);
     const toml = fs.readFileSync(path.join(root, '.codex', 'agents', 'code_auditor.toml'), 'utf8');
-    assert.match(toml, /Strict Semantic Review Attestation/);
-    assert.match(toml, /CAFEKIT_SEMANTIC_REVIEW_ATTESTATION/);
-    assert.match(toml, /MAC-protected host-hook observation/);
-    assert.doesNotMatch(toml, /host-signed/);
-    assert.match(toml, /specs\/<feature>\/spec\.json/);
+    assert.doesNotMatch(toml, /CAFEKIT_SEMANTIC_REVIEW_ATTESTATION/);
+    assert.doesNotMatch(toml, /Strict Semantic Review Attestation/);
+    assert.doesNotMatch(toml, /MAC-protected host-hook observation/);
   });
 });
 
@@ -3195,26 +3190,14 @@ test('Codex scaffold resolver rejects symlink template', () => {
   });
 });
 
-test('Claude and Codex installed Route preserve proportional live-catalog semantics', () => {
+test('Claude and Codex installed routing rules preserve the live-catalog semantics and install no Route', () => {
   inTempProject((root) => {
     const result = installPlatforms(root, ['claude', 'codex']);
     assert.equal(result.status, 0, `${result.stdout}\n${result.stderr}`);
-    const sourceRoot = path.join(PACKAGE_ROOT, 'src/claude/skills/route');
-    const expectedFiles = [
-      'SKILL.md', 'references/task-taxonomy.md',
-      'references/chaining-patterns.md', 'references/agent-timing.md'
-    ];
     const expectedRules = ['skill-workflow-routing.md', 'skill-domain-routing.md'];
     for (const runtime of ['claude', 'codex']) {
       const installedRoot = path.join(root, runtime === 'codex' ? '.agents/skills/route' : '.claude/skills/route');
-      for (const relative of expectedFiles) {
-        const source = fs.readFileSync(path.join(sourceRoot, relative), 'utf8');
-        const installed = fs.readFileSync(path.join(installedRoot, relative), 'utf8');
-        const expected = runtime === 'codex'
-          ? normalizeCodexBody(source, path.join(sourceRoot, relative))
-          : source;
-        assert.equal(installed, expected, `${runtime}:${relative}`);
-      }
+      assert.equal(fs.existsSync(installedRoot), false, `${runtime} still installs Route`);
       for (const relative of expectedRules) {
         const sourcePath = path.join(PACKAGE_ROOT, 'src/claude/rules', relative);
         const source = fs.readFileSync(sourcePath, 'utf8');
@@ -3224,12 +3207,8 @@ test('Claude and Codex installed Route preserve proportional live-catalog semant
         const expected = runtime === 'codex' ? normalizeCodexBody(source, sourcePath) : source;
         assert.equal(installed, expected, `${runtime}:rules/${relative}`);
       }
-      const route = fs.readFileSync(path.join(installedRoot, 'SKILL.md'), 'utf8');
-      assert.match(route, /names a valid installed skill[\s\S]*one installed skill clearly covers[\s\S]*Direct factual conversation/);
-      assert.match(route, /highest-link[\s\S]*risk[\s\S]*number of material domains/);
-      assert.match(route, /never expand it/);
       const catalog = installedCatalog(root, runtime);
-      assert.ok(catalog.skills.some((skill) => skill.public_id === 'cf:route'));
+      assert.equal(catalog.skills.some((skill) => skill.public_id === 'cf:route'), false);
       assert.equal(catalog.skills.some((skill) => skill.public_id === 'cf:docs'), false);
     }
   });
@@ -3304,24 +3283,3 @@ test('Claude and Codex installed rules preserve the ported review and process gu
   });
 });
 
-test('installed Route degrades safely when an agent is absent', () => {
-  inTempProject((root) => {
-    const result = installPlatforms(root, ['claude', 'codex']);
-    assert.equal(result.status, 0, `${result.stdout}\n${result.stderr}`);
-    const cases = [
-      ['claude', '.claude/agents/researcher.md', '.claude/skills/route'],
-      ['codex', '.codex/agents/researcher.toml', '.agents/skills/route'],
-    ];
-    for (const [runtime, agentRelative, routeRelative] of cases) {
-      fs.rmSync(path.join(root, agentRelative), { force: true });
-      assert.equal(fs.existsSync(path.join(root, agentRelative)), false);
-      const timing = fs.readFileSync(path.join(root, routeRelative, 'references/agent-timing.md'), 'utf8');
-      const route = fs.readFileSync(path.join(root, routeRelative, 'SKILL.md'), 'utf8');
-      const normalizedRoute = route.replace(/\s+/g, ' ');
-      assert.match(timing, /If the preferred agent is absent[\s\S]*never synthesize a role/);
-      assert.match(normalizedRoute, /continue inline when safe or return the named gap/);
-      assert.match(normalizedRoute, /Never synthesize an unavailable agent/);
-      assert.doesNotMatch(`${timing}\n${route}`, /researcher(?:\.md|\.toml)/i, runtime);
-    }
-  });
-});