forgedock 1.0.22 → 1.1.2

package/commands/pipeline-resume.md

@@ -1,6 +1,7 @@
 ---
 description: Recover full pipeline context after compaction — find the last active issue, reconstruct FORGE annotations, and resume work
 argument-hint: [issue number (optional — auto-detects if omitted)]
+install: extras
 ---
 
 # /pipeline-resume — Context Recovery After Compaction
@@ -9,8 +10,11 @@ argument-hint: [issue number (optional — auto-detects if omitted)]
 
 You are the context-recovery agent. When a session ends or context is compressed, all working state is lost. Your job is to restore it instantly — find the most recently active issue, read all FORGE annotations, reconstruct the current pipeline phase, and re-enter the pipeline exactly where it left off.
 
-**Agent model policy**: Default `model: "sonnet"`. Fallback: `model: "opus"` if Sonnet is rate-limited.
+**Agent model policy**: `model: "haiku"`, `effort: low` (mechanical tier — context recovery, annotation reading, state reconstruction). Fallback: `model: "sonnet"` if rate-limited. Feature gate: pass `effort` only on Claude Code >= 2.1.154.
 **NEVER use plan mode (EnterPlanMode).**
+**NEVER use the Agent tool** — pipeline-resume re-enters the pipeline via `Skill(skill="work-on", ...)` only. Using the Agent tool would restart the pipeline in an untracked subprocess rather than resuming the existing tracked session.
+
+<!-- FORGE:SPEC_LOADED — pipeline-resume.md loaded and active. Agent is bound by this spec. -->
 
 ---
 
@@ -36,6 +40,25 @@ If `forge.yaml` is missing: stop and tell the user to run `npx forgedock init` t
 
 Check `$ARGUMENTS`:
 
+- **`--all-stalled`** (fleet recovery mode): shell out to `npx forgedock resume-stalled` to scan all in-flight issues for expired leases and re-dispatch them. This delegates to the durable engine's `scanStalls` for ground-truth stall detection. Add `--dry-run` if only a report is needed. Then STOP — fleet recovery is complete.
+
+```bash
+# Fleet recovery mode — delegate to the engine's resume-stalled CLI
+if echo "$ARGUMENTS" | grep -q -- "--all-stalled"; then
+    DRY_RUN_FLAG=""
+    if echo "$ARGUMENTS" | grep -q -- "--dry-run"; then
+        DRY_RUN_FLAG="--dry-run"
+    fi
+    echo "Fleet stall recovery: scanning all in-flight issues for expired leases…"
+    npx forgedock resume-stalled $DRY_RUN_FLAG
+    echo ""
+    echo "Fleet recovery complete. To resume a single issue: /pipeline-resume <issue-number>"
+    # STOP — fleet recovery does not fall through to single-issue resume.
+fi
+```
+
+If `--all-stalled` was not specified, continue with single-issue resume:
+
 - **Explicit issue number** (e.g., `610`, `#610`): strip `#`, set `TARGET_NUMBER={number}`, skip to Phase 2.
 - **Empty**: proceed with auto-detection below.
 

package/commands/pipeline-status.md

@@ -1,6 +1,7 @@
 ---
 description: Pipeline-wide situational awareness — groups open issues by workflow state, shows active PRs, flags stale items, and reports milestone progress
 argument-hint: [--repo <owner/repo> | --stale-days <N>]
+install: extras
 ---
 
 # /pipeline-status — Pipeline Status
@@ -184,6 +185,70 @@ echo ""
 
 ---
 
+## Phase 3B: Engine Lease State (FORGE:STATE)
+
+For issues managed by the durable engine, read the `FORGE:STATE` block from each issue body and display lease status. This gives ground-truth stall detection beyond `updatedAt` heuristics. Gracefully degrades — issues without a `FORGE:STATE` block are skipped (they use the Phase 3 timestamp-based detection above).
+
+```bash
+echo "=== Engine Lease State ==="
+echo ""
+
+# gh --label uses AND filtering, so query each label separately and merge
+ENGINE_CANDIDATES=$(
+    for LABEL in "workflow:investigating" "workflow:ready-to-build" "workflow:building" "workflow:in-review"; do
+        gh issue list $GH_FLAG \
+            --state open \
+            --label "$LABEL" \
+            --limit 100 \
+            --json number,title,body 2>/dev/null || echo "[]"
+    done | jq -s 'flatten | unique_by(.number)'
+)
+
+NOW_MS=$(date +%s)000
+ENGINE_OUTPUT=""
+
+while IFS= read -r ROW; do
+    NUM=$(echo "$ROW" | jq -r '.number')
+    TITLE=$(echo "$ROW" | jq -r '.title[:55]')
+    BODY=$(echo "$ROW" | jq -r '.body // ""')
+
+    # Extract FORGE:STATE JSON block from issue body
+    STATE_JSON=$(echo "$BODY" | sed -n '/<!-- FORGE:STATE -->/,/<!-- \/FORGE:STATE -->/p' \
+        | grep -v '<!-- ' | tr -d '\n' 2>/dev/null || echo "")
+
+    if [ -z "$STATE_JSON" ]; then
+        continue  # No engine state — skip, Phase 3 covers this issue via updatedAt
+    fi
+
+    TERMINAL=$(echo "$STATE_JSON" | jq -r '.terminal // false' 2>/dev/null || echo "false")
+    if [ "$TERMINAL" = "true" ]; then
+        continue  # Terminal — not interesting for stall detection
+    fi
+
+    LEASE_UNTIL=$(echo "$STATE_JSON" | jq -r '.lease.until // 0' 2>/dev/null || echo "0")
+    LEASE_BY=$(echo "$STATE_JSON" | jq -r '.lease.by // "unknown"' 2>/dev/null || echo "unknown")
+    PHASE=$(echo "$STATE_JSON" | jq -r '.phase // "unknown"' 2>/dev/null || echo "unknown")
+
+    if [ "$LEASE_UNTIL" -gt "$NOW_MS" ] 2>/dev/null; then
+        REMAINING=$(( (LEASE_UNTIL - NOW_MS) / 60000 ))
+        ENGINE_OUTPUT="${ENGINE_OUTPUT}  #${NUM} ${TITLE} [${PHASE}] LEASED (${LEASE_BY}, ${REMAINING}m left)\n"
+    else
+        EXPIRED_AGO=$(( (NOW_MS - LEASE_UNTIL) / 60000 ))
+        ENGINE_OUTPUT="${ENGINE_OUTPUT}  #${NUM} ${TITLE} [${PHASE}] STALLED (lease expired ${EXPIRED_AGO}m ago)\n"
+    fi
+done <<< "$(echo "$ENGINE_CANDIDATES" | jq -c '.[]')"
+
+if [ -z "$ENGINE_OUTPUT" ]; then
+    echo "  No engine-managed issues in flight (or none with FORGE:STATE blocks)."
+else
+    echo -e "$ENGINE_OUTPUT"
+fi
+
+echo ""
+```
+
+---
+
 ## Phase 4: Milestone Progress
 
 Show open milestones with open/closed issue counts and percentage complete.

package/commands/qa-sweep.md

@@ -1,6 +1,7 @@
 ---
 description: Full platform QA sweep — auto-discovers every page, tests every UI element via browser automation, creates GitHub issues for all findings
 argument-hint: [all | dashboard | marketing | page <route> | journey | visual | functional | a11y]
+install: extras
 ---
 <!-- SPDX-FileCopyrightText: Copyright (c) RapierCraft Studios -->
 <!-- SPDX-License-Identifier: AGPL-3.0-or-later -->
@@ -11,9 +12,11 @@ argument-hint: [all | dashboard | marketing | page <route> | journey | visual |
 
 You are the QA orchestrator. Auto-discover every page across the platform (dashboard, marketing, blog, auth, docs, pricing), then systematically test every UI element, interaction, workflow, and state transition using browser automation. Create GitHub issues for every finding.
 
-**Agent model policy**: Default `model: "sonnet"`. If Sonnet is rate-limited, fall back to `model: "opus"`. User can override with `--model <name>`.
+**Agent model policy**: `model: "sonnet"` (standard tier). Fallback: `model: "opus"` if rate-limited. User can override with `--model <name>`. Feature gate: pass `effort` in Task/Skill spawns only on Claude Code >= 2.1.154.
 **NEVER use plan mode (EnterPlanMode).**
 
+<!-- FORGE:SPEC_LOADED — qa-sweep.md loaded and active. Agent is bound by this spec. -->
+
 ---
 
 ## Playwright MCP Prerequisite (MANDATORY — check before proceeding)