@ai-is-gonna/get-tasks-done 0.2.0

package/sdk/src/query/roadmap.ts

@@ -0,0 +1,894 @@
+/**
+ * Roadmap query handlers — ROADMAP.md analysis and phase lookup.
+ *
+ * Ported from get-tasks-done/bin/lib/roadmap.cjs and core.cjs.
+ * Provides roadmap.analyze (multi-pass parsing with disk correlation)
+ * and roadmap.get-phase (single phase section extraction).
+ *
+ * @example
+ * ```typescript
+ * import { roadmapAnalyze, roadmapGetPhase } from './roadmap.js';
+ *
+ * const analysis = await roadmapAnalyze([], '/project');
+ * // { data: { phases: [...], phase_count: 6, progress_percent: 50, ... } }
+ *
+ * const phase = await roadmapGetPhase(['10'], '/project');
+ * // { data: { found: true, phase_number: '10', phase_name: 'Read-Only Queries', ... } }
+ * ```
+ */
+
+import { existsSync } from 'node:fs';
+import { readFile, writeFile, readdir } from 'node:fs/promises';
+import { join } from 'node:path';
+import { GTDError, ErrorClassification } from '../errors.js';
+import { resolveGtdToolsPath } from '../sdk-package-compatibility.js';
+import {
+  escapeRegex,
+  normalizePhaseName,
+  phaseTokenMatches,
+  planningPaths,
+} from './helpers.js';
+import type { QueryHandler, QueryResult } from './utils.js';
+
+// ─── Internal types ───────────────────────────────────────────────────────
+
+interface PhaseSection {
+  found: boolean;
+  phase_number: string;
+  phase_name: string;
+  goal?: string | null;
+  /**
+   * Phase-level mode flag from `**Mode:** mvp` in ROADMAP.md.
+   * Lowercased + trimmed for canonical comparison; null when the field is absent.
+   * Unrecognized values are preserved verbatim for forward-compat (mirrors `roadmap.cjs`).
+   * Read by the `phase.mvp-mode` resolver and downstream MVP-aware workflows.
+   */
+  mode?: string | null;
+  success_criteria?: string[];
+  section?: string;
+  error?: string;
+  message?: string;
+}
+
+// ─── Exported helpers ─────────────────────────────────────────────────────
+
+/**
+ * Strip <details>...</details> blocks from content (shipped milestones).
+ *
+ * Port of stripShippedMilestones from core.cjs line 1082-1084.
+ */
+export function stripShippedMilestones(content: string): string {
+  // Pattern 1: <details>...</details> blocks (explicit collapse).
+  // <details\b[^>]*> tolerates attributes (e.g. <details open>, <details class="…">).
+  // Symmetry with extractCurrentMilestone()'s <details>-aware fallback (#2641):
+  // both functions must agree on what counts as a <details> opening tag, or
+  // shipped content wrapped in attributed tags would leak through here while
+  // the active-milestone anchor in extractCurrentMilestone() correctly fires.
+  let result = content.replace(/<details\b[^>]*>[\s\S]*?<\/details>/gi, '');
+  // Pattern 2: inline milestone headings marked as shipped.
+  // Keep aligned with heading levels accepted by extractCurrentMilestone() (## and ###).
+  const sections = result.split(/(?=^#{2,3}\s)/m);
+  result = sections.filter(s => !/^#{2,3}\s[^\n]*✅\s*SHIPPED\b/im.test(s)).join('');
+  return result;
+}
+
+/**
+ * Read milestone + name from STATE.md frontmatter when ROADMAP does not encode them.
+ */
+async function parseMilestoneFromState(projectDir: string, workstream?: string): Promise<{ version: string; name: string } | null> {
+  try {
+    const stateRaw = await readFile(planningPaths(projectDir, workstream).state, 'utf-8');
+    const vm = stateRaw.match(/^milestone:\s*(.+)$/m);
+    if (!vm) return null;
+    const version = vm[1].trim().replace(/^["']|["']$/g, '');
+    const nm = stateRaw.match(/^milestone_name:\s*(.+)$/m);
+    const name = nm ? nm[1].trim().replace(/^["']|["']$/g, '') : 'milestone';
+    return { version, name };
+  } catch {
+    return null;
+  }
+}
+
+/**
+ * Get milestone version and name from ROADMAP.md (and optionally STATE.md).
+ *
+ * Port of getMilestoneInfo from core.cjs lines 1367-1402, extended for:
+ * - 🟡 in-flight marker (same list shape as 🚧)
+ * - milestone bullets `**vX.Y Title**` before `## Phases` (last = current when listed in semver order)
+ * - STATE.md frontmatter when ROADMAP has no parseable milestone
+ * - **last** bare `vX.Y` fallback (first match was often v1.0 from the shipped list)
+ *
+ * @param projectDir - Project root directory
+ * @returns Object with version and name
+ */
+export async function getMilestoneInfo(projectDir: string, workstream?: string): Promise<{ version: string; name: string }> {
+  try {
+    // Priority 1: STATE.md frontmatter (authoritative for version; name only when real)
+    const fromState = await parseMilestoneFromState(projectDir, workstream);
+    const stateVersion = fromState?.version ?? null;
+    const stateName = fromState && fromState.name !== 'milestone' ? fromState.name : null;
+    if (stateVersion && stateName) {
+      return { version: stateVersion, name: stateName };
+    }
+    // STATE.md has a version but no real name — fall through to ROADMAP for the name,
+    // then override the version with the authoritative STATE.md value.
+
+    const roadmap = await readFile(planningPaths(projectDir, workstream).roadmap, 'utf-8');
+
+    // List-format: construction / blocked (legacy emoji)
+    const barricadeMatch = roadmap.match(/🚧\s*\*\*v(\d+(?:\.\d+)+)\s+([^*]+)\*\*/);
+    if (barricadeMatch) {
+      return { version: stateVersion ?? 'v' + barricadeMatch[1], name: barricadeMatch[2].trim() };
+    }
+
+    // List-format: in flight / active (GTD ROADMAP template uses 🟡 for current milestone)
+    const inFlightMatch = roadmap.match(/🟡\s*\*\*v(\d+(?:\.\d+)+)\s+([^*]+)\*\*/);
+    if (inFlightMatch) {
+      return { version: stateVersion ?? 'v' + inFlightMatch[1], name: inFlightMatch[2].trim() };
+    }
+
+    // Heading-format — strip shipped <details> blocks first
+    const cleaned = stripShippedMilestones(roadmap);
+    const headingMatch = cleaned.match(/##\s+.*v(\d+(?:\.\d+)+)[:\s]+([^\n(]+)/);
+    if (headingMatch) {
+      return { version: stateVersion ?? 'v' + headingMatch[1], name: headingMatch[2].trim() };
+    }
+
+    // Milestone bullet list (## Milestones … ## Phases): use last **vX.Y Title** — typically the current row
+    const beforePhases = roadmap.split(/^##\s+Phases\b/m)[0] ?? roadmap;
+    const boldMatches = [...beforePhases.matchAll(/\*\*v(\d+(?:\.\d+)+)\s+([^*]+)\*\*/g)];
+    if (boldMatches.length > 0) {
+      const last = boldMatches[boldMatches.length - 1];
+      return { version: stateVersion ?? 'v' + last[1], name: last[2].trim() };
+    }
+
+    const allBare = [...cleaned.matchAll(/\bv(\d+(?:\.\d+)+)\b/g)];
+    if (allBare.length > 0) {
+      const lastBare = allBare[allBare.length - 1];
+      return { version: stateVersion ?? lastBare[0], name: 'milestone' };
+    }
+
+    return { version: stateVersion ?? 'v1.0', name: 'milestone' };
+  } catch {
+    const fromState = await parseMilestoneFromState(projectDir, workstream);
+    if (fromState) return fromState;
+    return { version: 'v1.0', name: 'milestone' };
+  }
+}
+
+/**
+ * Extract the current milestone section from ROADMAP.md.
+ *
+ * Two anchoring strategies, tried in order:
+ *   1. Markdown heading containing the active version (`^#{1,3}\s+.*vX.Y…`).
+ *   2. `<details><summary>vX.Y…</summary>…</details>` block (the GitHub-friendly
+ *      collapse pattern; see #2641). When this fallback fires, the captured
+ *      `<summary>` text is synthesized as a `##` heading prepended to the
+ *      returned slice so downstream consumers that scan for milestone headings
+ *      (e.g. the `data.milestones` loop in `roadmapAnalyze`) still see an
+ *      active-milestone anchor.
+ *
+ * If neither strategy matches the active version, falls through to
+ * `stripShippedMilestones(content)`.
+ *
+ * Originally ported from core.cjs lines 1102-1170; the TS implementation has
+ * since diverged (Backlog-leak fix #2422, phase-vX.Y truncation fix #2619,
+ * fenced-code-block tracking #2787, `<details><summary>` fallback #2641).
+ *
+ * @param content - Full ROADMAP.md content
+ * @param projectDir - Working directory for reading STATE.md
+ * @returns Content scoped to current milestone
+ */
+export async function extractCurrentMilestone(content: string, projectDir: string, workstream?: string): Promise<string> {
+  // Get version from STATE.md frontmatter.
+  // Strip optional surrounding YAML quotes (e.g. `milestone: "v0.9"`) for parity
+  // with parseMilestoneFromState() above and getMilestoneInfo()'s STATE.md path.
+  // Without this, a quoted version yields `escapedVersion = '\\"v0\\.9\\"'`
+  // which matches neither markdown headings nor <summary> text, falling
+  // through to stripShippedMilestones() — and reintroducing the same archived-
+  // milestone misrouting this fallback addresses.
+  let version: string | null = null;
+  try {
+    const stateRaw = await readFile(planningPaths(projectDir, workstream).state, 'utf-8');
+    const milestoneMatch = stateRaw.match(/^milestone:\s*(.+)/m);
+    if (milestoneMatch) {
+      version = milestoneMatch[1].trim().replace(/^["']|["']$/g, '');
+    }
+  } catch { /* intentionally empty */ }
+
+  // Fallback: derive from ROADMAP in-progress marker
+  if (!version) {
+    const inProgressMatch = content.match(/(?:🚧|🟡)\s*\*\*v(\d+(?:\.\d+)+)\s/);
+    if (inProgressMatch) {
+      version = 'v' + inProgressMatch[1];
+    }
+  }
+
+  if (!version) return stripShippedMilestones(content);
+
+  // Find section matching this version
+  const escapedVersion = escapeRegex(version);
+  const sectionPattern = new RegExp(
+    `(^#{1,3}\\s+.*${escapedVersion}(?![\\d.])[^\\n]*)`,
+    'mi'
+  );
+  const sectionMatch = content.match(sectionPattern);
+
+  if (!sectionMatch || sectionMatch.index === undefined) {
+    // Fallback: <details><summary> matching the active version (issue #2641).
+    //
+    // Many projects (GitHub-friendly collapse pattern) wrap the active
+    // milestone's phase details inside a collapsible block whose <summary>
+    // names the version, e.g.:
+    //
+    //   <details>
+    //   <summary>v0.9 Local-First Bus (active) — Phase Details</summary>
+    //   ### Phase 1: ...
+    //   </details>
+    //
+    // The markdown-heading lookup above misses this because <summary> is HTML,
+    // not a heading. Without this fallback, control falls through to
+    // stripShippedMilestones() which removes ALL <details> blocks
+    // indiscriminately — including the active milestone's — causing
+    // roadmapGetPhase() to return {found:false} for phases that ARE in the
+    // active ROADMAP. The init.phase-op safety guard then misfires and can
+    // route phase lookups into archived milestones.
+    //
+    // Regex anatomy:
+    //   <details\b[^>]*>          tolerate attributes (e.g. <details open>)
+    //   \s*<summary\b[^>]*>       tolerate attributes on <summary>
+    //   ((?:(?!</summary>).)*?    non-greedy summary capture; tolerates
+    //     ${escapedVersion}        inline HTML in the summary text
+    //     (?![\d.])                non-version-character lookahead — prevents
+    //                                 `v0.1` from substring-matching `v0.10`
+    //     (?:(?!</summary>).)*)
+    //   </summary>                 end of summary
+    //   ([\s\S]*?)</details>       lazy body capture to the FIRST </details>
+    //
+    // Contract: any consumer that scans the returned slice for milestone
+    // headings (e.g. /##\s*.*vX.Y/) sees the active milestone's anchor. We
+    // synthesize that heading from the captured <summary> text rather than
+    // returning the body alone.
+    //
+    // Hardening guards:
+    //   - Nested <details>: the lazy quantifier truncates at the inner
+    //     </details>, silently losing trailing phases. Detect and fall through
+    //     to stripShippedMilestones() instead of returning truncated content.
+    //   - Empty body: a <details> block with no body would synthesize a heading
+    //     with nothing under it. Treat as no-match.
+    //   - Summary sanitization: strip inline HTML (e.g. <em>active</em>) and
+    //     leading `#` tokens before promoting to a `##` heading, so the result
+    //     is a single well-formed markdown heading.
+    const detailsPattern = new RegExp(
+      `<details\\b[^>]*>\\s*<summary\\b[^>]*>` +
+      `((?:(?!</summary>).)*?${escapedVersion}(?![\\d.])(?:(?!</summary>).)*)` +
+      `</summary>([\\s\\S]*?)</details>`,
+      'i'
+    );
+    const detailsMatch = content.match(detailsPattern);
+    if (
+      detailsMatch &&
+      detailsMatch[2].trim() &&                    // empty-body guard
+      !detailsMatch[2].includes('<details')        // nested-<details> guard
+    ) {
+      const summary = detailsMatch[1]
+        .replace(/<[^>]+>/g, '')                   // strip inline HTML
+        .replace(/^#+\s*/, '')                     // strip leading `#`
+        .trim();
+      const body = detailsMatch[2];
+      return `## ${summary}\n${body}`;
+    }
+    return stripShippedMilestones(content);
+  }
+
+  const sectionStart = sectionMatch.index;
+
+  // Find end: next milestone heading at same or higher level, or EOF.
+  // Skip headings that belong to the SAME version (e.g. "## v2.0 Phase Details").
+  const headingLevelMatch = sectionMatch[1].match(/^(#{1,3})\s/);
+  const headingLevel = headingLevelMatch ? headingLevelMatch[1].length : 2;
+  const restContent = content.slice(sectionStart + sectionMatch[0].length);
+  // Extract current version so same-version sub-headings are not treated as boundaries.
+  // Capture full semver (major.minor.patch) so v2.0.1 is not collapsed to "2.0".
+  const currentVersionMatch = version ? version.match(/v(\d+(?:\.\d+)+)/i) : null;
+  const currentVersionStr = currentVersionMatch ? currentVersionMatch[1] : '';
+
+  // Exclude phase headings (e.g. "### Phase 12: v1.0 Tech-Debt Closure") from
+  // being treated as milestone boundaries just because they mention vX.Y in
+  // the title. Phase headings always start with the literal `Phase `. See #2619.
+  const nextMilestoneRegex = new RegExp(
+    `^#{1,${headingLevel}}\\s+(?!Phase\\s+\\S)(?:.*v(\\d+(?:\\.\\d+)+)[^\\n]*|.*(?:✅|📋|🚧|🟡))`,
+    // `i` flag ensures the `(?!Phase\s+\S)` lookahead matches PHASE/phase too
+    // (CodeRabbit follow-up on #2619).
+    'gmi'
+  );
+
+  let sectionEnd = content.length;
+  let m: RegExpExecArray | null;
+  while ((m = nextMilestoneRegex.exec(restContent)) !== null) {
+    const matchedVersion = m[1];
+    // Skip headings that reference the same version (e.g. "## v2.0 Phase Details").
+    if (matchedVersion && currentVersionStr && matchedVersion === currentVersionStr) continue;
+    sectionEnd = sectionStart + sectionMatch[0].length + m.index;
+    break;
+  }
+
+  // Issue #3493: a generic, non-version-prefixed `## Phase Details` (or
+  // `### Phase Details`) heading sometimes sits AFTER a planned-milestone
+  // sibling (e.g. `### 📋 v2.1+ (Planned)`) in document order but holds the
+  // detail bodies (`### Phase N: ...`) for the *active* milestone. Issue #2422
+  // / PR #2455 handled the version-prefixed variant `## v2.0 Phase Details`
+  // via the same-version `continue` above; this branch handles the generic-
+  // label variant. When such a heading is found after the initial sectionEnd,
+  // append the Phase Details block to the returned slice (skipping the
+  // intervening planned-milestone content so it does not leak in), stopping
+  // at the next real milestone boundary (version-bearing or milestone-emoji-
+  // bearing heading) or EOF.
+  //
+  // Bounded to a single append so a malformed roadmap can't loop. Only
+  // matches the literal `Phase Details` label (canonical per GTD ROADMAP
+  // template); anything else continues to terminate the slice.
+  let phaseDetailsTail = '';
+  if (sectionEnd < content.length) {
+    const afterEnd = content.slice(sectionEnd);
+    const genericPhaseDetailsRegex = /^(#{1,3})\s+Phase\s+Details\b[^\n]*$/im;
+    const phaseDetailsMatch = afterEnd.match(genericPhaseDetailsRegex);
+    if (phaseDetailsMatch && phaseDetailsMatch.index !== undefined) {
+      const pdStart = sectionEnd + phaseDetailsMatch.index;
+      const pdHeadingLen = phaseDetailsMatch[0].length;
+      const pdLevel = phaseDetailsMatch[1].length;
+      // Scan from after the Phase Details heading for the next real milestone
+      // boundary (different version OR milestone emoji). Reuses the same
+      // (?!Phase\s+\S) phase-heading guard from #2619, plus a
+      // (?!Phase\s+Details\b) guard so a sibling Phase Details heading
+      // doesn't terminate the block prematurely.
+      const afterPdHeading = content.slice(pdStart + pdHeadingLen);
+      const nextRealMilestoneRegex = new RegExp(
+        `^#{1,${pdLevel}}\\s+(?!Phase\\s+\\S)(?!Phase\\s+Details\\b)(?:.*v(\\d+(?:\\.\\d+)+)[^\\n]*|.*(?:✅|📋|🚧|🟡))`,
+        'gmi'
+      );
+      let pdEnd = content.length;
+      let mm: RegExpExecArray | null;
+      while ((mm = nextRealMilestoneRegex.exec(afterPdHeading)) !== null) {
+        const mv = mm[1];
+        if (mv && currentVersionStr && mv === currentVersionStr) continue;
+        pdEnd = pdStart + pdHeadingLen + mm.index;
+        break;
+      }
+      phaseDetailsTail = '\n' + content.slice(pdStart, pdEnd);
+    }
+  }
+
+  // Return only the current milestone section — never include the preamble, which
+  // may contain ## Backlog and other non-current-milestone phases. Append any
+  // generic Phase Details block discovered above (#3493) so detail-section
+  // lookups for the active milestone succeed.
+  return content.slice(sectionStart, sectionEnd) + phaseDetailsTail;
+}
+
+// ─── Next-milestone helpers (issue #2497) ─────────────────────────────────
+
+/**
+ * Phase shape returned by extractPhasesFromSection — mirrors the fields used
+ * by the current-milestone phases array in initManager so consumers can
+ * render queued phases uniformly.
+ */
+export interface QueuedPhase {
+  number: string;
+  name: string;
+  goal: string | null;
+  depends_on: string | null;
+}
+
+/**
+ * Extract phase entries from an arbitrary ROADMAP milestone section.
+ *
+ * Parses `#### Phase N: Name` / `### Phase N: Name` / `## Phase N: Name`
+ * headings and, for each, captures goal + depends_on via the same patterns
+ * used by initManager's current-milestone phase parsing. Used by
+ * `initManager` to populate `queued_phases` (#2497).
+ */
+export function extractPhasesFromSection(section: string): QueuedPhase[] {
+  const phasePattern = /#{2,4}\s*Phase\s+(\d+[A-Z]?(?:\.\d+)*)\s*:\s*([^\n]+)/gi;
+  const phases: QueuedPhase[] = [];
+  let m: RegExpExecArray | null;
+  while ((m = phasePattern.exec(section)) !== null) {
+    const phaseNum = m[1];
+    const phaseName = m[2].replace(/\(INSERTED\)/i, '').trim();
+    const sectionStart = m.index;
+    const rest = section.slice(sectionStart);
+    const nextHeader = rest.match(/\n#{2,4}\s+Phase\s+\d/i);
+    const end = nextHeader ? sectionStart + (nextHeader.index ?? 0) : section.length;
+    const body = section.slice(sectionStart, end);
+    const goalMatch = body.match(/\*\*Goal(?::\*\*|\*\*:)\s*([^\n]+)/i);
+    const dependsMatch = body.match(/\*\*Depends on(?::\*\*|\*\*:)\s*([^\n]+)/i);
+    phases.push({
+      number: phaseNum,
+      name: phaseName,
+      goal: goalMatch ? goalMatch[1].trim() : null,
+      depends_on: dependsMatch ? dependsMatch[1].trim() : null,
+    });
+  }
+  return phases;
+}
+
+/**
+ * Find the milestone section that comes immediately AFTER the active one.
+ *
+ * Used by initManager to surface `queued_phases` without conflating the
+ * active milestone's phase list with the next one (#2497). Returns null
+ * when no subsequent milestone section exists (active is the last one).
+ *
+ * Reuses the same current-version resolution path as `getMilestoneInfo`:
+ * STATE.md frontmatter first, then in-flight emoji markers in ROADMAP.
+ * Shipped milestones are stripped first so they can't shadow the real
+ * "next" one.
+ */
+export async function extractNextMilestoneSection(
+  content: string,
+  projectDir: string,
+): Promise<{ version: string; name: string; section: string } | null> {
+  const cleaned = stripShippedMilestones(content);
+
+  // Resolve current version via STATE.md (priority) then in-flight markers.
+  let currentVersion: string | null = null;
+  const fromState = await parseMilestoneFromState(projectDir);
+  if (fromState?.version) {
+    const raw = fromState.version.trim();
+    currentVersion = /^v\d/i.test(raw) ? raw : `v${raw}`;
+  }
+  if (!currentVersion) {
+    const inProgressMatch = cleaned.match(/(?:🚧|🟡)\s*\*\*v(\d+(?:\.\d+)+)\s/);
+    if (inProgressMatch) currentVersion = 'v' + inProgressMatch[1];
+  }
+  if (!currentVersion) return null;
+
+  // Find the current milestone ## heading.
+  const escaped = escapeRegex(currentVersion);
+  const currentHeadingPattern = new RegExp(
+    `^##\\s+[^\\n]*${escaped}[^\\n]*$`,
+    'mi',
+  );
+  const currentMatch = cleaned.match(currentHeadingPattern);
+  if (!currentMatch || currentMatch.index === undefined) return null;
+
+  // Look for the next ## milestone heading after the current one.
+  const tail = cleaned.slice(currentMatch.index + currentMatch[0].length);
+  // Exclude phase headings — see #2619.
+  const nextMilestonePattern = /^##\s+(?!Phase\s+\S)([^\n]*(?:v(\d+(?:\.\d+)+)|✅|🚧|🟡|📋)[^\n]*)$/gim;
+  let nextMatch: RegExpExecArray | null;
+  while ((nextMatch = nextMilestonePattern.exec(tail)) !== null) {
+    const heading = nextMatch[1];
+    const versionMatch = heading.match(/v(\d+(?:\.\d+)+)/);
+    if (!versionMatch) continue;
+    const nextVersion = 'v' + versionMatch[1];
+    if (nextVersion === currentVersion) continue;
+
+    // Derive a display name: trim through "vX.Y:" or "vX.Y —" prefix.
+    const nameMatch = heading.match(/v\d+(?:\.\d+)+:?\s*[—–-]?\s*([^\n(]+)/);
+    const name = nameMatch ? nameMatch[1].trim() : heading.trim();
+
+    const sectionStart = (nextMatch.index ?? 0) + nextMatch[0].length;
+    const afterStart = tail.slice(sectionStart);
+    const followingHeader = afterStart.match(/^##\s/m);
+    const sectionEnd = followingHeader && followingHeader.index !== undefined
+      ? sectionStart + followingHeader.index
+      : tail.length;
+    const section = tail.slice(sectionStart, sectionEnd);
+
+    return { version: nextVersion, name, section };
+  }
+
+  return null;
+}
+
+// ─── Internal helpers ─────────────────────────────────────────────────────
+
+/**
+ * Search for a phase section in roadmap content.
+ *
+ * Port of searchPhaseInContent from roadmap.cjs lines 14-73.
+ */
+function searchPhaseInContent(content: string, escapedPhase: string, phaseNum: string): PhaseSection | null {
+  // Match "## Phase X:", "### Phase X:", or "#### Phase X:" with optional name
+  const phasePattern = new RegExp(
+    `#{2,4}\\s*Phase\\s+${escapedPhase}:\\s*([^\\n]+)`,
+    'i'
+  );
+  const headerMatch = content.match(phasePattern);
+
+  if (!headerMatch) {
+    // Fallback: check if phase exists in summary list but missing detail section
+    const checklistPattern = new RegExp(
+      `-\\s*\\[[ x]\\]\\s*\\*\\*Phase\\s+${escapedPhase}:\\s*([^*]+)\\*\\*`,
+      'i'
+    );
+    const checklistMatch = content.match(checklistPattern);
+
+    if (checklistMatch) {
+      return {
+        found: false,
+        phase_number: phaseNum,
+        phase_name: checklistMatch[1].trim(),
+        error: 'malformed_roadmap',
+        message: `Phase ${phaseNum} exists in summary list but missing "### Phase ${phaseNum}:" detail section. ROADMAP.md needs both formats.`,
+      };
+    }
+
+    return null;
+  }
+
+  const phaseName = headerMatch[1].trim();
+  const headerIndex = headerMatch.index!;
+
+  // Find the end of this section (next ## or ### phase header, or end of file)
+  const restOfContent = content.slice(headerIndex);
+  const nextHeaderMatch = restOfContent.match(/\n#{2,4}\s+Phase\s+\d/i);
+  const sectionEnd = nextHeaderMatch
+    ? headerIndex + nextHeaderMatch.index!
+    : content.length;
+
+  const section = content.slice(headerIndex, sectionEnd).trim();
+
+  // Extract goal if present (supports both **Goal:** and **Goal**: formats)
+  const goalMatch = section.match(/\*\*Goal(?::\*\*|\*\*:)\s*([^\n]+)/i);
+  const goal = goalMatch ? goalMatch[1].trim() : null;
+
+  // Mode: vertical-MVP slice mode flag. Lowercased + trimmed for canonical
+  // comparison; unrecognized values preserved verbatim for forward-compat.
+  // Mirrors roadmap.cjs:120-123 — restoring parity that was missed in the SDK port.
+  const modeMatch = section.match(/\*\*Mode(?::\*\*|\*\*:)\s*([^\n]+)/i);
+  const mode = modeMatch ? modeMatch[1].trim().toLowerCase() : null;
+
+  // Extract success criteria as structured array
+  const criteriaMatch = section.match(/\*\*Success Criteria\*\*[^\n]*:\s*\n((?:\s*\d+\.\s*[^\n]+\n?)+)/i);
+  const success_criteria = criteriaMatch
+    ? criteriaMatch[1].trim().split('\n').map(line => line.replace(/^\s*\d+\.\s*/, '').trim()).filter(Boolean)
+    : [];
+
+  return {
+    found: true,
+    phase_number: phaseNum,
+    phase_name: phaseName,
+    goal,
+    mode,
+    success_criteria,
+    section,
+  };
+}
+
+async function countPhasePlansAndSummaries(phaseDir: string): Promise<{ planCount: number; summaryCount: number; hasContext: boolean; hasResearch: boolean; }> {
+  const phaseFiles = await readdir(phaseDir);
+  const rootPlans = phaseFiles.filter(f => f.endsWith('-PLAN.md') || f === 'PLAN.md');
+  const rootSummaries = phaseFiles.filter(f => f.endsWith('-SUMMARY.md') || f === 'SUMMARY.md');
+
+  let nestedPlans: string[] = [];
+  let nestedSummaries: string[] = [];
+  const plansDir = join(phaseDir, 'plans');
+  if (existsSync(plansDir)) {
+    const files = await readdir(plansDir);
+    nestedPlans = files.filter(f => /^PLAN-\d+.*\.md$/i.test(f));
+    nestedSummaries = files.filter(f => /^SUMMARY-\d+.*\.md$/i.test(f));
+  }
+
+  return {
+    planCount: rootPlans.length + nestedPlans.length,
+    summaryCount: rootSummaries.length + nestedSummaries.length,
+    hasContext: phaseFiles.some(f => f.endsWith('-CONTEXT.md') || f === 'CONTEXT.md'),
+    hasResearch: phaseFiles.some(f => f.endsWith('-RESEARCH.md') || f === 'RESEARCH.md'),
+  };
+}
+
+// ─── Exported handlers ────────────────────────────────────────────────────
+
+/**
+ * Query handler for roadmap.get-phase.
+ *
+ * Port of cmdRoadmapGetPhase from roadmap.cjs lines 75-113.
+ *
+ * @param args - args[0] is phase number (required)
+ * @param projectDir - Project root directory
+ * @returns QueryResult with phase section info or { found: false }
+ */
+export const roadmapGetPhase: QueryHandler = async (args, projectDir, workstream) => {
+  const phaseNum = args[0];
+  if (!phaseNum) {
+    throw new GTDError(
+      'Usage: roadmap get-phase <phase-number>',
+      ErrorClassification.Validation,
+    );
+  }
+
+  const roadmapPath = planningPaths(projectDir, workstream).roadmap;
+
+  let rawContent: string;
+  try {
+    rawContent = await readFile(roadmapPath, 'utf-8');
+  } catch {
+    return { data: { found: false, error: 'ROADMAP.md not found' } };
+  }
+
+  const milestoneContent = await extractCurrentMilestone(rawContent, projectDir, workstream);
+  const escapedPhase = escapeRegex(phaseNum);
+
+  // Search the current milestone slice first, then fall back to full roadmap.
+  const fullContent = stripShippedMilestones(rawContent);
+  const milestoneResult = searchPhaseInContent(milestoneContent, escapedPhase, phaseNum);
+  const result = (milestoneResult && !milestoneResult.error)
+    ? milestoneResult
+    : searchPhaseInContent(fullContent, escapedPhase, phaseNum) || milestoneResult;
+
+  if (!result) {
+    return { data: { found: false, phase_number: phaseNum } };
+  }
+
+  return { data: result };
+};
+
+/**
+ * Query handler for roadmap.analyze.
+ *
+ * Port of cmdRoadmapAnalyze from roadmap.cjs lines 115-248.
+ * Multi-pass regex parsing with disk status correlation.
+ *
+ * @param args - Unused
+ * @param projectDir - Project root directory
+ * @returns QueryResult with full roadmap analysis
+ */
+export const roadmapAnalyze: QueryHandler = async (_args, projectDir, workstream) => {
+  const roadmapPath = planningPaths(projectDir, workstream).roadmap;
+
+  let rawContent: string;
+  try {
+    rawContent = await readFile(roadmapPath, 'utf-8');
+  } catch {
+    return { data: { error: 'ROADMAP.md not found', milestones: [], phases: [], current_phase: null } };
+  }
+
+  const content = await extractCurrentMilestone(rawContent, projectDir, workstream);
+  const phasesDir = planningPaths(projectDir, workstream).phases;
+
+  // IMPORTANT: Create regex INSIDE the function to avoid /g lastIndex persistence
+  const phasePattern = /#{2,4}\s*Phase\s+(\d+[A-Z]?(?:\.\d+)*)\s*:\s*([^\n]+)/gi;
+  const phases: Array<Record<string, unknown>> = [];
+  let match: RegExpExecArray | null;
+
+  while ((match = phasePattern.exec(content)) !== null) {
+    const phaseNum = match[1];
+    const phaseName = match[2].replace(/\(INSERTED\)/i, '').trim();
+
+    // Extract goal from the section
+    const sectionStart = match.index;
+    const restOfContent = content.slice(sectionStart);
+    const nextHeader = restOfContent.match(/\n#{2,4}\s+Phase\s+\d/i);
+    const sectionEnd = nextHeader ? sectionStart + nextHeader.index! : content.length;
+    const section = content.slice(sectionStart, sectionEnd);
+
+    const goalMatch = section.match(/\*\*Goal(?::\*\*|\*\*:)\s*([^\n]+)/i);
+    const goal = goalMatch ? goalMatch[1].trim() : null;
+
+    const dependsMatch = section.match(/\*\*Depends on(?::\*\*|\*\*:)\s*([^\n]+)/i);
+    const depends_on = dependsMatch ? dependsMatch[1].trim() : null;
+
+    // Check completion on disk
+    const normalized = normalizePhaseName(phaseNum);
+    let diskStatus = 'no_directory';
+    let planCount = 0;
+    let summaryCount = 0;
+    let hasContext = false;
+    let hasResearch = false;
+
+    try {
+      const entries = await readdir(phasesDir, { withFileTypes: true });
+      const dirs = entries.filter(e => e.isDirectory()).map(e => e.name);
+      const dirMatch = dirs.find(d => phaseTokenMatches(d, normalized));
+
+      if (dirMatch) {
+        const counts = await countPhasePlansAndSummaries(join(phasesDir, dirMatch));
+        planCount = counts.planCount;
+        summaryCount = counts.summaryCount;
+        hasContext = counts.hasContext;
+        hasResearch = counts.hasResearch;
+
+        if (summaryCount >= planCount && planCount > 0) diskStatus = 'complete';
+        else if (summaryCount > 0) diskStatus = 'partial';
+        else if (planCount > 0) diskStatus = 'planned';
+        else if (hasResearch) diskStatus = 'researched';
+        else if (hasContext) diskStatus = 'discussed';
+        else diskStatus = 'empty';
+      }
+    } catch { /* intentionally empty */ }
+
+    // Check ROADMAP checkbox status
+    const checkboxPattern = new RegExp(`-\\s*\\[(x| )\\]\\s*.*Phase\\s+${escapeRegex(phaseNum)}[:\\s]`, 'i');
+    const checkboxMatch = content.match(checkboxPattern);
+    const roadmapComplete = checkboxMatch ? checkboxMatch[1] === 'x' : false;
+
+    // If roadmap marks phase complete, trust that over disk
+    if (roadmapComplete && diskStatus !== 'complete') {
+      diskStatus = 'complete';
+    }
+
+    phases.push({
+      number: phaseNum,
+      name: phaseName,
+      goal,
+      depends_on,
+      plan_count: planCount,
+      summary_count: summaryCount,
+      has_context: hasContext,
+      has_research: hasResearch,
+      disk_status: diskStatus,
+      roadmap_complete: roadmapComplete,
+    });
+  }
+
+  // Extract milestone info
+  const milestones: Array<{ heading: string; version: string }> = [];
+  const milestonePattern = /##\s*(.*v(\d+(?:\.\d+)+)[^(\n]*)/gi;
+  let mMatch: RegExpExecArray | null;
+  while ((mMatch = milestonePattern.exec(content)) !== null) {
+    milestones.push({
+      heading: mMatch[1].trim(),
+      version: 'v' + mMatch[2],
+    });
+  }
+
+  // Find current and next phase
+  const currentPhase = phases.find(p => p.disk_status === 'planned' || p.disk_status === 'partial') || null;
+  const nextPhase = phases.find(p => p.disk_status === 'empty' || p.disk_status === 'no_directory' || p.disk_status === 'discussed' || p.disk_status === 'researched') || null;
+
+  // Aggregated stats
+  const totalPlans = phases.reduce((sum, p) => sum + (p.plan_count as number), 0);
+  const totalSummaries = phases.reduce((sum, p) => sum + (p.summary_count as number), 0);
+  const completedPhases = phases.filter(p => p.disk_status === 'complete').length;
+
+  // Detect phases in summary list without detail sections (malformed ROADMAP)
+  const checklistPattern = /-\s*\[[ x]\]\s*\*\*Phase\s+(\d+[A-Z]?(?:\.\d+)*)/gi;
+  const checklistPhases = new Set<string>();
+  let checklistMatch: RegExpExecArray | null;
+  while ((checklistMatch = checklistPattern.exec(content)) !== null) {
+    checklistPhases.add(checklistMatch[1]);
+  }
+  const detailPhases = new Set(phases.map(p => p.number as string));
+  const missingDetails = [...checklistPhases].filter(p => !detailPhases.has(p));
+
+  const result: Record<string, unknown> = {
+    milestones,
+    phases,
+    phase_count: phases.length,
+    completed_phases: completedPhases,
+    total_plans: totalPlans,
+    total_summaries: totalSummaries,
+    progress_percent: totalPlans > 0 ? Math.min(100, Math.round((totalSummaries / totalPlans) * 100)) : 0,
+    current_phase: currentPhase ? currentPhase.number : null,
+    next_phase: nextPhase ? nextPhase.number : null,
+    missing_phase_details: missingDetails.length > 0 ? missingDetails : null,
+  };
+
+  return { data: result };
+};
+
+
+// ─── roadmapAnnotateDependencies ─────────────────────────────────────────
+
+/**
+ * Annotate the ROADMAP.md plan list with wave dependency notes and
+ * cross-cutting constraints derived from PLAN frontmatter.
+ *
+ * Delegates to gtd-tools.cjs which holds the full annotation logic.
+ * Returns { updated, phase, waves, cross_cutting_constraints }.
+ */
+export const roadmapAnnotateDependencies: QueryHandler = async (args, projectDir) => {
+  const phase = args[0];
+  if (!phase) {
+    return { data: { updated: false, reason: 'phase argument required' } };
+  }
+
+  const { spawnSync } = await import('node:child_process');
+  const toolsPath = resolveGtdToolsPath(projectDir);
+
+  const result = spawnSync(process.execPath, [toolsPath, 'roadmap', 'annotate-dependencies', phase], {
+    cwd: projectDir,
+    encoding: 'utf-8',
+    stdio: ['pipe', 'pipe', 'pipe'],
+    timeout: 15000,
+    maxBuffer: 1024 * 1024,
+  });
+
+  if (result.error) {
+    return { data: { updated: false, reason: result.error.message || 'gtd-tools invocation failed' } };
+  }
+
+  if (result.status !== 0) {
+    return { data: { updated: false, reason: result.stderr?.trim() || 'gtd-tools error' } };
+  }
+
+  try {
+    return { data: JSON.parse(result.stdout.trim()) };
+  } catch {
+    return { data: { updated: false, reason: 'failed to parse gtd-tools output' } };
+  }
+};
+
+
+// ─── requirementsMarkComplete ─────────────────────────────────────────────
+
+/**
+ * Mark requirement IDs complete in REQUIREMENTS.md (checkbox + traceability table).
+ * Port of `cmdRequirementsMarkComplete` from milestone.cjs lines 11–87.
+ */
+export const requirementsMarkComplete: QueryHandler = async (args, projectDir, workstream) => {
+  if (args.length === 0) {
+    throw new GTDError(
+      'requirement IDs required. Usage: requirements mark-complete REQ-01,REQ-02 or REQ-01 REQ-02',
+      ErrorClassification.Validation,
+    );
+  }
+
+  const reqIds = args
+    .join(' ')
+    .replace(/[\[\]]/g, '')
+    .split(/[,\s]+/)
+    .map(r => r.trim())
+    .filter(Boolean);
+
+  if (reqIds.length === 0) {
+    throw new GTDError('no valid requirement IDs found', ErrorClassification.Validation);
+  }
+
+  const paths = planningPaths(projectDir, workstream);
+  if (!existsSync(paths.requirements)) {
+    return { data: { updated: false, reason: 'REQUIREMENTS.md not found', ids: reqIds } };
+  }
+
+  let reqContent = (await readFile(paths.requirements, 'utf-8')).replace(/\r\n/g, '\n');
+  const updated: string[] = [];
+  const alreadyComplete: string[] = [];
+  const notFound: string[] = [];
+
+  for (const reqId of reqIds) {
+    let found = false;
+    const reqEscaped = escapeRegex(reqId);
+
+    const checkboxPattern = new RegExp(`(-\\s*\\[)[ ](\\]\\s*\\*\\*${reqEscaped}\\*\\*)`, 'gi');
+    const afterCheckbox = reqContent.replace(checkboxPattern, '$1x$2');
+    if (afterCheckbox !== reqContent) {
+      reqContent = afterCheckbox;
+      found = true;
+    }
+
+    const tablePattern = new RegExp(`(\\|\\s*${reqEscaped}\\s*\\|[^|]+\\|)\\s*Pending\\s*(\\|)`, 'gi');
+    const afterTable = reqContent.replace(tablePattern, '$1 Complete $2');
+    if (afterTable !== reqContent) {
+      reqContent = afterTable;
+      found = true;
+    }
+
+    if (found) {
+      updated.push(reqId);
+    } else {
+      const doneCheckbox = new RegExp(`-\\s*\\[x\\]\\s*\\*\\*${reqEscaped}\\*\\*`, 'i');
+      const doneTable = new RegExp(`\\|\\s*${reqEscaped}\\s*\\|[^|]+\\|\\s*Complete\\s*\\|`, 'i');
+      if (doneCheckbox.test(reqContent) || doneTable.test(reqContent)) {
+        alreadyComplete.push(reqId);
+      } else {
+        notFound.push(reqId);
+      }
+    }
+  }
+
+  if (updated.length > 0) {
+    await writeFile(paths.requirements, reqContent, 'utf-8');
+  }
+
+  return {
+    data: {
+      updated: updated.length > 0,
+      marked_complete: updated,
+      already_complete: alreadyComplete,
+      not_found: notFound,
+      total: reqIds.length,
+    },
+  };
+};