crewly 1.14.1 → 1.16.0

package/config/skills/agent/core/resolve-prediction/execute.test.sh

@@ -0,0 +1,74 @@
+#!/bin/bash
+# Tests for resolve-prediction — run with: bash execute.test.sh
+# Spins up a python HTTP stub as the backend, asserts the request path/body
+# the skill sends and the JSON it prints. Exit 0 on pass.
+set -u
+HERE="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+EXEC="$HERE/execute.sh"
+PASS=0; FAIL=0
+check() {
+  local name="$1"; local got="$2"; local want="$3"
+  if [ "$got" = "$want" ]; then PASS=$((PASS+1)); else FAIL=$((FAIL+1)); echo "FAIL: $name"; echo "  want: $want"; echo "  got:  $got"; fi
+}
+
+PORT=18794
+STUB_LOG="$(mktemp)"; export STUB_LOG; export STUB_PORT="$PORT"
+python3 -u <<'PY' >/dev/null 2>&1 &
+from http.server import BaseHTTPRequestHandler, HTTPServer
+import json, os
+LOG = os.environ['STUB_LOG']; PORT = int(os.environ['STUB_PORT'])
+class H(BaseHTTPRequestHandler):
+    def do_POST(self):
+        n = int(self.headers.get('content-length', '0')); body = self.rfile.read(n).decode() if n else ''
+        data = json.loads(body or '{}')
+        open(LOG, 'w').write(json.dumps({'path': self.path, 'body': data}))
+        self.send_header_json = lambda code: (self.send_response(code), self.send_header('Content-Type', 'application/json'), self.end_headers())
+        if '/pred-missing/' in self.path:
+            self.send_header_json(404); self.wfile.write(b'{"success":false,"error":"Prediction pred-missing not found"}'); return
+        if 'accurate' not in data and data.get('outcome') not in ('correct', 'wrong'):
+            self.send_header_json(400); self.wfile.write(b'{"success":false,"error":"accurate (boolean) is required"}'); return
+        acc = data.get('accurate', data.get('outcome') == 'correct')
+        pred = {'id': 'pred-1-abc', 'prediction': 'x', 'confidence': 0.9, 'madeAt': '2026-09-18', 'outcome': data.get('outcome'), 'accurate': acc, 'resolvedAt': '2026-09-18'}
+        self.send_header_json(200)
+        self.wfile.write(json.dumps({'success': True, 'data': {'prediction': pred, 'calibrationScore': 0.82}}).encode())
+    def log_message(self, *a, **k): pass
+HTTPServer(('127.0.0.1', PORT), H).serve_forever()
+PY
+STUB_PID=$!
+disown "$STUB_PID" 2>/dev/null || true
+trap 'kill "$STUB_PID" >/dev/null 2>&1; rm -f "$STUB_LOG"' EXIT
+for i in $(seq 1 30); do curl -s -o /dev/null -m 0.2 "http://127.0.0.1:${PORT}/" 2>/dev/null && break; sleep 0.1; done
+
+run() { CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="crewly-dev-1" bash "$EXEC" "$@" 2>/dev/null; }
+run_err() { CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="" bash "$EXEC" "$@" 2>&1 >/dev/null | jq -r '.error' 2>/dev/null; }
+
+OUT=$(run --id pred-1-abc --outcome "merged Thursday" --accurate true)
+check "flags: success + calibration" "$(printf '%s' "$OUT" | jq -c '{success, calibrationScore}')" '{"success":true,"calibrationScore":0.82}'
+check "flags: prediction accurate" "$(printf '%s' "$OUT" | jq -r .prediction.accurate)" "true"
+check "flags: path includes id" "$(jq -r .path "$STUB_LOG")" "/api/agents/crewly-dev-1/self-improvement/predictions/pred-1-abc/resolve"
+check "flags: body (accurate is boolean)" "$(jq -c .body "$STUB_LOG")" '{"outcome":"merged Thursday","accurate":true}'
+
+run --id pred-1-abc --outcome wrong >/dev/null
+check "verdict-only outcome omits accurate" "$(jq -c .body "$STUB_LOG")" '{"outcome":"wrong"}'
+
+run '{"id":"pred-1-abc","outcome":"slipped","accurate":false,"sessionName":"crewly-qa-9"}' >/dev/null
+check "json: body" "$(jq -c .body "$STUB_LOG")" '{"outcome":"slipped","accurate":false}'
+check "json: explicit session in path" "$(jq -r .path "$STUB_LOG")" "/api/agents/crewly-qa-9/self-improvement/predictions/pred-1-abc/resolve"
+
+run '{"predictionId":"pred-1-abc","outcome":"correct"}' >/dev/null
+check "json: accepts legacy predictionId key" "$(jq -r .path "$STUB_LOG")" "/api/agents/crewly-dev-1/self-improvement/predictions/pred-1-abc/resolve"
+
+OUT=$(CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="crewly-dev-1" bash "$EXEC" --id pred-missing --outcome correct 2>/dev/null); RC=$?
+check "backend 404: exit code" "$RC" "1"
+check "backend 404: reason" "$(printf '%s' "$OUT" | jq -r .reason)" "Prediction pred-missing not found"
+
+OUT=$(CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="crewly-dev-1" bash "$EXEC" --id pred-1-abc --outcome "took a while" 2>/dev/null); RC=$?
+check "backend 400 (ambiguous outcome): reason" "$(printf '%s' "$OUT" | jq -r .reason)" "accurate (boolean) is required"
+
+check "bad accurate value rejected locally" "$(run_err --session s --id p --outcome x --accurate maybe)" "accurate must be true or false (got: maybe)"
+check "missing id errors" "$(run_err --session s --outcome x)" "Missing required parameter: id (--id)"
+check "missing outcome errors" "$(run_err --session s --id p)" "Missing required parameter: outcome (--outcome)"
+check "missing session errors" "$(run_err --id p --outcome x)" "Missing required parameter: sessionName (--session or CREWLY_SESSION_NAME)"
+
+echo "=== Results: $PASS passed, $FAIL failed ==="
+[ "$FAIL" -eq 0 ]

package/config/skills/agent/core/set-focus/SKILL.md

@@ -0,0 +1,64 @@
+---
+name: Set Focus
+description: Replace your current focus list — the topics you should prioritise until further notice. Injected into your prompt on the next session.
+version: 1.0.0
+category: memory
+skillType: claude-skill
+assignableRoles:
+  - developer
+  - qa
+  - tpm
+  - designer
+  - frontend-developer
+  - backend-developer
+  - fullstack-dev
+  - qa-engineer
+  - product-manager
+  - architect
+  - generalist
+  - sales
+  - support
+  - orchestrator
+triggers:
+  - set focus
+  - focus on
+  - prioritise
+  - prioritize
+tags:
+  - memory
+  - attention
+  - self-improvement
+execution:
+  type: script
+  script:
+    file: execute.sh
+    interpreter: bash
+    timeoutMs: 15000
+---
+
+# Set Focus
+
+Replace your focus list with a new set of topics. Focus items are the things you should keep front of mind — an active incident, a deadline, a recurring mistake you are correcting. They show up in your prompt (up to 5) and are pruned automatically after two weeks without an update, so refresh them when they still matter.
+
+Setting focus **replaces** the whole list. To drop a single distraction instead, use `suppress-noise`.
+
+## Parameters
+
+| Flag | JSON Field | Required | Description |
+|------|-----------|----------|-------------|
+| `--item` / `-i` (repeatable) | `items` (array) | Yes | Focus items, one per flag (or a JSON array) |
+| `--session` / `-s` | `sessionName` | No | Your session name — defaults to `$CREWLY_SESSION_NAME` |
+
+## Examples
+
+```bash
+# CLI flags (preferred)
+bash execute.sh --item "ship checkout v2 by Friday" --item "flaky payment e2e"
+
+# Legacy JSON
+bash execute.sh '{"items":["ship checkout v2 by Friday","flaky payment e2e"]}'
+```
+
+## Output
+
+`{ "success": true, "focus": [...], "suppressed": [...] }` — the attention state after the update.

package/config/skills/agent/core/set-focus/execute.sh

@@ -0,0 +1,73 @@
+#!/bin/bash
+# =============================================================================
+# set-focus — Replace the agent's focus list (attention management)
+#
+# Backed by POST /api/agents/:sessionName/self-improvement/attention/focus.
+# Focus items are injected into the agent's prompt (≤5) and pruned after two
+# weeks without an update.
+#
+# Usage:
+#   bash execute.sh --item "ship v2" --item "flaky CI" [--session crewly-dev-1]
+#   bash execute.sh '{"items":["ship v2","flaky CI"],"sessionName":"crewly-dev-1"}'
+# =============================================================================
+set -euo pipefail
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+source "${SCRIPT_DIR}/../../_common/lib.sh"
+
+print_usage() {
+  cat <<'EOF_USAGE'
+Usage:
+  bash execute.sh --item "ship v2" --item "flaky CI" [--session crewly-dev-1]
+  bash execute.sh '{"items":["ship v2","flaky CI"]}'
+
+Options:
+  --item    | -i   Focus item (repeat for several; at least one required)
+  --session | -s   Session name (defaults to $CREWLY_SESSION_NAME)
+  --help    | -h   Show this help
+EOF_USAGE
+}
+
+INPUT_JSON=""
+SESSION_NAME=""
+ITEMS_JSON='[]'
+
+if [[ $# -gt 0 && ${1:0:1} == '{' ]]; then
+  INPUT_JSON="$1"
+  shift || true
+fi
+
+while [[ $# -gt 0 ]]; do
+  case "$1" in
+    --item|-i)
+      [ $# -ge 2 ] || error_exit "--item requires a value"
+      ITEMS_JSON=$(jq -cn --argjson arr "$ITEMS_JSON" --arg v "$2" '$arr + [$v]')
+      shift 2 ;;
+    --session|-s)
+      [ $# -ge 2 ] || error_exit "--session requires a value"
+      SESSION_NAME="$2"; shift 2 ;;
+    --help|-h) print_usage; exit 0 ;;
+    *) error_exit "Unknown option: $1" ;;
+  esac
+done
+
+if [ -n "$INPUT_JSON" ]; then
+  INPUT=$(read_json_input "$INPUT_JSON")
+  [ -z "$SESSION_NAME" ] && SESSION_NAME=$(printf '%s' "$INPUT" | jq -r '.sessionName // .agentId // empty')
+  if [ "$(printf '%s' "$ITEMS_JSON" | jq 'length')" -eq 0 ]; then
+    ITEMS_JSON=$(printf '%s' "$INPUT" | jq -c '(.items // []) | if type == "string" then [.] else . end')
+  fi
+fi
+
+[ -z "$SESSION_NAME" ] && SESSION_NAME="${CREWLY_SESSION_NAME:-}"
+require_param "sessionName (--session or CREWLY_SESSION_NAME)" "$SESSION_NAME"
+[ "$(printf '%s' "$ITEMS_JSON" | jq 'length')" -gt 0 ] || error_exit "Missing required parameter: items (--item)"
+
+BODY=$(jq -cn --argjson items "$ITEMS_JSON" '{items: $items}')
+
+RESPONSE=$(api_call POST "/agents/${SESSION_NAME}/self-improvement/attention/focus" "$BODY" 2>&1) || {
+  ERROR_MSG=$(printf '%s' "$RESPONSE" | jq -r '.details.error // .details // .error // "Unknown error"' 2>/dev/null || echo "$RESPONSE")
+  jq -n --arg reason "$ERROR_MSG" '{success: false, reason: $reason}'
+  exit 1
+}
+
+printf '%s' "$RESPONSE" | jq -c '{success: (.success // false), focus: (.data.focus // []), suppressed: (.data.suppressed // [])}'

package/config/skills/agent/core/set-focus/execute.test.sh

@@ -0,0 +1,67 @@
+#!/bin/bash
+# Tests for set-focus — run with: bash execute.test.sh
+# Spins up a python HTTP stub as the backend, asserts the request path/body
+# the skill sends and the JSON it prints. Exit 0 on pass.
+set -u
+HERE="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+EXEC="$HERE/execute.sh"
+PASS=0; FAIL=0
+check() {
+  local name="$1"; local got="$2"; local want="$3"
+  if [ "$got" = "$want" ]; then PASS=$((PASS+1)); else FAIL=$((FAIL+1)); echo "FAIL: $name"; echo "  want: $want"; echo "  got:  $got"; fi
+}
+
+PORT=18791
+STUB_LOG="$(mktemp)"; export STUB_LOG; export STUB_PORT="$PORT"
+python3 -u <<'PY' >/dev/null 2>&1 &
+from http.server import BaseHTTPRequestHandler, HTTPServer
+import json, os
+LOG = os.environ['STUB_LOG']; PORT = int(os.environ['STUB_PORT'])
+class H(BaseHTTPRequestHandler):
+    def do_POST(self):
+        n = int(self.headers.get('content-length', '0')); body = self.rfile.read(n).decode() if n else ''
+        open(LOG, 'w').write(json.dumps({'path': self.path, 'body': json.loads(body or '{}')}))
+        items = json.loads(body).get('items', [])
+        if '__reject__' in items:
+            self.send_response(400); self.send_header('Content-Type', 'application/json'); self.end_headers()
+            self.wfile.write(b'{"success":false,"error":"items must be an array of non-empty strings"}'); return
+        self.send_response(200); self.send_header('Content-Type', 'application/json'); self.end_headers()
+        self.wfile.write(json.dumps({'success': True, 'data': {'focus': items, 'suppressed': ['old']}}).encode())
+    def log_message(self, *a, **k): pass
+HTTPServer(('127.0.0.1', PORT), H).serve_forever()
+PY
+STUB_PID=$!
+disown "$STUB_PID" 2>/dev/null || true
+trap 'kill "$STUB_PID" >/dev/null 2>&1; rm -f "$STUB_LOG"' EXIT
+for i in $(seq 1 30); do curl -s -o /dev/null -m 0.2 "http://127.0.0.1:${PORT}/" 2>/dev/null && break; sleep 0.1; done
+
+run() { CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="crewly-dev-1" bash "$EXEC" "$@" 2>/dev/null; }
+run_err() { CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="" bash "$EXEC" "$@" 2>&1 >/dev/null | jq -r '.error' 2>/dev/null; }
+
+# --- flags ---
+OUT=$(run --item "ship v2" --item "flaky CI")
+check "flags: output" "$OUT" '{"success":true,"focus":["ship v2","flaky CI"],"suppressed":["old"]}'
+check "flags: path uses CREWLY_SESSION_NAME" "$(jq -r .path "$STUB_LOG")" "/api/agents/crewly-dev-1/self-improvement/attention/focus"
+check "flags: body" "$(jq -c .body "$STUB_LOG")" '{"items":["ship v2","flaky CI"]}'
+
+# --- legacy JSON with explicit session ---
+OUT=$(run '{"items":["a"],"sessionName":"crewly-qa-9"}')
+check "json: output" "$OUT" '{"success":true,"focus":["a"],"suppressed":["old"]}'
+check "json: explicit session in path" "$(jq -r .path "$STUB_LOG")" "/api/agents/crewly-qa-9/self-improvement/attention/focus"
+
+# --- json string coerced to array ---
+run '{"items":"solo"}' >/dev/null
+check "json: string items coerced to array" "$(jq -c .body "$STUB_LOG")" '{"items":["solo"]}'
+
+# --- backend error surfaces as success:false and exit 1 ---
+OUT=$(CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="crewly-dev-1" bash "$EXEC" --item __reject__ 2>/dev/null); RC=$?
+check "backend 400: exit code" "$RC" "1"
+check "backend 400: reason" "$(printf '%s' "$OUT" | jq -r .reason)" "items must be an array of non-empty strings"
+
+# --- validation ---
+check "missing items errors" "$(run_err --session s1)" "Missing required parameter: items (--item)"
+check "missing session errors" "$(run_err --item x)" "Missing required parameter: sessionName (--session or CREWLY_SESSION_NAME)"
+check "unknown option errors" "$(run_err --bogus 1)" "Unknown option: --bogus"
+
+echo "=== Results: $PASS passed, $FAIL failed ==="
+[ "$FAIL" -eq 0 ]

package/config/skills/agent/core/suppress-noise/SKILL.md

@@ -0,0 +1,59 @@
+---
+name: Suppress Noise
+description: Mark a topic as noise so you stop giving it attention. Suppressed topics are listed in your prompt and removed from your focus list.
+version: 1.0.0
+category: memory
+skillType: claude-skill
+assignableRoles:
+  - developer
+  - qa
+  - tpm
+  - designer
+  - frontend-developer
+  - backend-developer
+  - fullstack-dev
+  - qa-engineer
+  - product-manager
+  - architect
+  - generalist
+  - sales
+  - support
+  - orchestrator
+triggers:
+  - suppress
+  - ignore topic
+  - stop tracking
+  - that's noise
+tags:
+  - memory
+  - attention
+  - self-improvement
+execution:
+  type: script
+  script:
+    file: execute.sh
+    interpreter: bash
+    timeoutMs: 15000
+---
+
+# Suppress Noise
+
+Add one topic to your suppressed list. Use it when past context keeps pulling you toward something that no longer matters (a retired service, a resolved incident, a stale request). The item is removed from your focus list if it was there, and shows up in your prompt under "Ignore" (up to 5).
+
+## Parameters
+
+| Flag | JSON Field | Required | Description |
+|------|-----------|----------|-------------|
+| `--item` / `-i` | `item` | Yes | The topic to suppress |
+| `--session` / `-s` | `sessionName` | No | Your session name — defaults to `$CREWLY_SESSION_NAME` |
+
+## Examples
+
+```bash
+bash execute.sh --item "legacy webhook retries"
+bash execute.sh '{"item":"legacy webhook retries"}'
+```
+
+## Output
+
+`{ "success": true, "focus": [...], "suppressed": [...] }` — the attention state after the update.

package/config/skills/agent/core/suppress-noise/execute.sh

@@ -0,0 +1,68 @@
+#!/bin/bash
+# =============================================================================
+# suppress-noise — Add a topic to the agent's suppressed list
+#
+# Backed by POST /api/agents/:sessionName/self-improvement/attention/suppress.
+#
+# Usage:
+#   bash execute.sh --item "legacy webhooks" [--session crewly-dev-1]
+#   bash execute.sh '{"item":"legacy webhooks","sessionName":"crewly-dev-1"}'
+# =============================================================================
+set -euo pipefail
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+source "${SCRIPT_DIR}/../../_common/lib.sh"
+
+print_usage() {
+  cat <<'EOF_USAGE'
+Usage:
+  bash execute.sh --item "legacy webhooks" [--session crewly-dev-1]
+  bash execute.sh '{"item":"legacy webhooks"}'
+
+Options:
+  --item    | -i   Topic to suppress (required)
+  --session | -s   Session name (defaults to $CREWLY_SESSION_NAME)
+  --help    | -h   Show this help
+EOF_USAGE
+}
+
+INPUT_JSON=""
+SESSION_NAME=""
+ITEM=""
+
+if [[ $# -gt 0 && ${1:0:1} == '{' ]]; then
+  INPUT_JSON="$1"
+  shift || true
+fi
+
+while [[ $# -gt 0 ]]; do
+  case "$1" in
+    --item|-i)
+      [ $# -ge 2 ] || error_exit "--item requires a value"
+      ITEM="$2"; shift 2 ;;
+    --session|-s)
+      [ $# -ge 2 ] || error_exit "--session requires a value"
+      SESSION_NAME="$2"; shift 2 ;;
+    --help|-h) print_usage; exit 0 ;;
+    *) error_exit "Unknown option: $1" ;;
+  esac
+done
+
+if [ -n "$INPUT_JSON" ]; then
+  INPUT=$(read_json_input "$INPUT_JSON")
+  [ -z "$SESSION_NAME" ] && SESSION_NAME=$(printf '%s' "$INPUT" | jq -r '.sessionName // .agentId // empty')
+  [ -z "$ITEM" ] && ITEM=$(printf '%s' "$INPUT" | jq -r '.item // empty')
+fi
+
+[ -z "$SESSION_NAME" ] && SESSION_NAME="${CREWLY_SESSION_NAME:-}"
+require_param "sessionName (--session or CREWLY_SESSION_NAME)" "$SESSION_NAME"
+require_param "item (--item)" "$ITEM"
+
+BODY=$(jq -cn --arg item "$ITEM" '{item: $item}')
+
+RESPONSE=$(api_call POST "/agents/${SESSION_NAME}/self-improvement/attention/suppress" "$BODY" 2>&1) || {
+  ERROR_MSG=$(printf '%s' "$RESPONSE" | jq -r '.details.error // .details // .error // "Unknown error"' 2>/dev/null || echo "$RESPONSE")
+  jq -n --arg reason "$ERROR_MSG" '{success: false, reason: $reason}'
+  exit 1
+}
+
+printf '%s' "$RESPONSE" | jq -c '{success: (.success // false), focus: (.data.focus // []), suppressed: (.data.suppressed // [])}'

package/config/skills/agent/core/suppress-noise/execute.test.sh

@@ -0,0 +1,58 @@
+#!/bin/bash
+# Tests for suppress-noise — run with: bash execute.test.sh
+# Spins up a python HTTP stub as the backend, asserts the request path/body
+# the skill sends and the JSON it prints. Exit 0 on pass.
+set -u
+HERE="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+EXEC="$HERE/execute.sh"
+PASS=0; FAIL=0
+check() {
+  local name="$1"; local got="$2"; local want="$3"
+  if [ "$got" = "$want" ]; then PASS=$((PASS+1)); else FAIL=$((FAIL+1)); echo "FAIL: $name"; echo "  want: $want"; echo "  got:  $got"; fi
+}
+
+PORT=18792
+STUB_LOG="$(mktemp)"; export STUB_LOG; export STUB_PORT="$PORT"
+python3 -u <<'PY' >/dev/null 2>&1 &
+from http.server import BaseHTTPRequestHandler, HTTPServer
+import json, os
+LOG = os.environ['STUB_LOG']; PORT = int(os.environ['STUB_PORT'])
+class H(BaseHTTPRequestHandler):
+    def do_POST(self):
+        n = int(self.headers.get('content-length', '0')); body = self.rfile.read(n).decode() if n else ''
+        data = json.loads(body or '{}')
+        open(LOG, 'w').write(json.dumps({'path': self.path, 'body': data}))
+        if data.get('item') == '__reject__':
+            self.send_response(400); self.send_header('Content-Type', 'application/json'); self.end_headers()
+            self.wfile.write(b'{"success":false,"error":"item must be a non-empty string"}'); return
+        self.send_response(200); self.send_header('Content-Type', 'application/json'); self.end_headers()
+        self.wfile.write(json.dumps({'success': True, 'data': {'focus': ['keep'], 'suppressed': [data.get('item')]}}).encode())
+    def log_message(self, *a, **k): pass
+HTTPServer(('127.0.0.1', PORT), H).serve_forever()
+PY
+STUB_PID=$!
+disown "$STUB_PID" 2>/dev/null || true
+trap 'kill "$STUB_PID" >/dev/null 2>&1; rm -f "$STUB_LOG"' EXIT
+for i in $(seq 1 30); do curl -s -o /dev/null -m 0.2 "http://127.0.0.1:${PORT}/" 2>/dev/null && break; sleep 0.1; done
+
+run() { CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="crewly-dev-1" bash "$EXEC" "$@" 2>/dev/null; }
+run_err() { CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="" bash "$EXEC" "$@" 2>&1 >/dev/null | jq -r '.error' 2>/dev/null; }
+
+OUT=$(run --item "legacy webhooks")
+check "flags: output" "$OUT" '{"success":true,"focus":["keep"],"suppressed":["legacy webhooks"]}'
+check "flags: path" "$(jq -r .path "$STUB_LOG")" "/api/agents/crewly-dev-1/self-improvement/attention/suppress"
+check "flags: body" "$(jq -c .body "$STUB_LOG")" '{"item":"legacy webhooks"}'
+
+OUT=$(run '{"item":"noise (with) \"quotes\"","sessionName":"crewly-qa-9"}')
+check "json: special characters survive" "$(jq -r .body.item "$STUB_LOG")" 'noise (with) "quotes"'
+check "json: explicit session in path" "$(jq -r .path "$STUB_LOG")" "/api/agents/crewly-qa-9/self-improvement/attention/suppress"
+
+OUT=$(CREWLY_API_URL="http://127.0.0.1:${PORT}" CREWLY_SESSION_NAME="crewly-dev-1" bash "$EXEC" --item __reject__ 2>/dev/null); RC=$?
+check "backend 400: exit code" "$RC" "1"
+check "backend 400: reason" "$(printf '%s' "$OUT" | jq -r .reason)" "item must be a non-empty string"
+
+check "missing item errors" "$(run_err --session s1)" "Missing required parameter: item (--item)"
+check "missing session errors" "$(run_err --item x)" "Missing required parameter: sessionName (--session or CREWLY_SESSION_NAME)"
+
+echo "=== Results: $PASS passed, $FAIL failed ==="
+[ "$FAIL" -eq 0 ]

package/config/skills/orchestrator/delegate-task/SKILL.md

@@ -47,7 +47,7 @@ The script auto-resolves `config/skills/...` references to absolute paths so del
 | `--task-type` | `taskType` | No | Task type: `general`, `technical` (default: `general`) |
 | `--force-cross-team` | `forceCrossTeam` | No | Allow cross-team delegation |
 | `--request-id` / `-R` | `requestId` | No | Parent V3 Request ID; tags the resulting WorkItem so `Request.workItemIds[]` is populated by the SLA subscriber (Pipeline-#4 fix) |
-| `monitor` (JSON only) | `monitor` | No | Auto-monitoring config (see below) |
+| `--fallback-minutes` | — | No | Minutes until the ONE §3.0 fallback check fires for this delegation (default `120` ≈ 2× a TL milestone ETA). `0` disables it. Not cancelled on completion — keep it at 2× ETA, never a poll interval |
 
 ## Usage — CLI Flags (preferred)
 
@@ -71,51 +71,47 @@ bash execute.sh --to agent-joe --task-file /tmp/task-description.txt --priority
 bash execute.sh '{"to":"agent-joe","task":"Implement the login form","priority":"high","context":"Use React hooks","projectPath":"/path/to/project"}'
 ```
 
-### With monitoring disabled (opt-out, JSON only)
+### With the fallback timer disabled (opt-out)
 
 ```bash
-bash execute.sh '{"to":"agent-joe","task":"Implement the login form","priority":"high","projectPath":"/path/to/project","monitor":{"idleEvent":false,"fallbackCheckMinutes":0}}'
+bash execute.sh --to agent-joe --task "Implement the login form" --priority high --project /path/to/project --fallback-minutes 0
 ```
 
 ## Examples
 
-### Example 1: Basic delegation (monitoring enabled by default)
+### Example 1: Basic delegation (one fallback timer by default)
 ```bash
 bash config/skills/orchestrator/delegate-task/execute.sh '{"to":"agent-joe","task":"Fix the login bug","priority":"high"}'
 ```
 This will automatically:
-1. Send the task to agent-joe's terminal
-2. Subscribe to `agent:idle` for agent-joe (auto-notifies when agent goes idle)
-3. Schedule a recurring check every 5 minutes (auto-reminds orchestrator to check progress)
-4. Link all monitoring IDs to the task for auto-cleanup on completion
+1. Create + claim the WorkItem in the task pool and deliver it to agent-joe's terminal
+2. Schedule ONE fallback check (`fallbackTriggerId`, fires after `--fallback-minutes`, default 120) that wakes the orchestrator with a "Fallback check on agent-joe" WorkItem if the work is still open
 
-### Example 2: Delegation with project tracking + monitoring
+It does NOT set up a recurring check — the reconciler escalates stalled or unverified work on its own. Add the `agent:idle_after_task` watch yourself per §3.0; never add a second fallback or a self-targeted "check on agent" WorkItem for the same delegation.
+
+### Example 2: Delegation with project tracking
 ```bash
 bash config/skills/orchestrator/delegate-task/execute.sh '{"to":"agent-joe","task":"Implement user auth","priority":"high","projectPath":"/path/to/project"}'
 ```
 Also creates a task file in the project's `.crewly/tasks/` directory.
 
-### Example 3: Delegation with custom monitoring interval
+### Example 3: Cross-team delegation with a longer fallback (2× an 8 h ETA)
 ```bash
-bash config/skills/orchestrator/delegate-task/execute.sh '{"to":"agent-sam","task":"Write unit tests","priority":"normal","monitor":{"fallbackCheckMinutes":10}}'
+bash config/skills/orchestrator/delegate-task/execute.sh --to agent-sam --task "Ship the billing rewrite" --priority normal --fallback-minutes 720
 ```
 
-### Example 4: Delegation with monitoring disabled
+### Example 4: Quick fix with the fallback disabled
 ```bash
-bash config/skills/orchestrator/delegate-task/execute.sh '{"to":"agent-joe","task":"Quick fix","priority":"low","monitor":{"idleEvent":false,"fallbackCheckMinutes":0}}'
+bash config/skills/orchestrator/delegate-task/execute.sh --to agent-joe --task "Quick fix" --priority low --fallback-minutes 0
 ```
 
 ## Output
 
 JSON confirmation of task delivery. When `projectPath` is provided, also returns the created task file path.
 
-## Auto-Cleanup
-
-When the agent completes the task (via `report-status` with `status: done` or `complete-task`), all linked monitoring is automatically cleaned up:
-- Scheduled checks are cancelled
-- Event subscriptions are unsubscribed
+## Cleanup
 
-No manual cleanup needed.
+The fallback trigger is **not** cancelled when the task completes. When it fires on already-verified work, complete the resulting check WorkItem immediately; to avoid the wake-up altogether, `cancel-followup --id <fallbackTriggerId from this skill's output>` once you have verified the deliverable (§3.0 step 3).
 
 ## Delivery Strategy
 
@@ -141,6 +137,6 @@ Error messages are output to **stdout** (JSON format) so the orchestrator can re
 
 - `assign-task` — for formal task tracking in the management system (file-based kanban)
 - `send-message` — for simple messages without task structure
-- `subscribe-event` — manual event subscription (auto-handled when using `monitor`)
-- `schedule-check` — manual schedule creation (auto-handled when using `monitor`)
+- `watch-for-event` — the `agent:idle_after_task` watch that completes the §3.0 loop (not created by this skill)
+- `cancel-followup` — cancel the fallback trigger once the deliverable is verified
 - `report-status` — agent reports completion, triggers auto-cleanup

package/config/skills/orchestrator/delegate-task/execute.sh

@@ -17,16 +17,25 @@ TASK_TYPE="general"
 TEAM_ID=""
 FORCE_CROSS_TEAM="false"
 REQUEST_ID=""
-# Self-heal fix #3 (2026-05-20): default fallback timer at 30min. Per
-# the orchestrator prompt §3.0 the orc is supposed to set 2× ETA
-# manually for every dispatch, but the convention was unenforced —
-# observable in the 2026-05-20 ESTestNode incident where the orc set
-# only the idle subscription and not the fallback timer, leaving the
-# orc clueless when Sora never started. Auto-creating a 30min fallback
-# at dispatch time guarantees the dual-signal contract is honored.
+# Self-heal fix #3 (2026-05-20): default fallback timer. Per the
+# orchestrator prompt §3.0 the orc is supposed to set 2× ETA manually
+# for every dispatch, but the convention was unenforced — observable in
+# the 2026-05-20 ESTestNode incident where the orc set only the idle
+# subscription and not the fallback timer, leaving the orc clueless when
+# Sora never started. Auto-creating the fallback at dispatch time
+# guarantees the dual-signal contract is honored.
+#
+# Default raised 30 → 120 min (2026-09-18): the trigger is never cancelled
+# on completion, so at 30 min it fired inside the normal 30–90 min TL
+# ETA on essentially every delegation and materialised a "Fallback check
+# on <agent>" WorkItem for the orc — 26 of 65 real WorkItems on one
+# production box, each a full-context wake-up that found nothing to do.
+# 120 min is the §3.0 "2× ETA" figure for TL milestone delegations; the
+# reconciler (2h unverified escalation, task:queued_too_long) covers the
+# genuinely stuck case before this fires.
 # Caller can override via --fallback-minutes N, or disable with
 # --fallback-minutes 0.
-FALLBACK_MINUTES="30"
+FALLBACK_MINUTES="120"
 
 # Detect legacy JSON argument
 if [[ $# -gt 0 && ${1:0:1} == '{' ]]; then
@@ -345,7 +354,7 @@ if [ -n "$WI_ID" ] && [ "${FALLBACK_MINUTES:-0}" -gt 0 ] 2>/dev/null; then
       --arg createdBy "delegate-task" \
       --arg name "fallback-${TO}-${WI_ID:0:8}" \
       --arg title "Fallback check on ${TO} for task ${WI_ID:0:8}" \
-      --arg description "Per §3.0: ${FALLBACK_MINUTES}min fallback after dispatch to ${TO}. Check whether ${WI_ID} is making progress — if running but no output, re-prompt; if still queued, re-target or fail-fast." \
+      --arg description "Per §3.0: ${FALLBACK_MINUTES}min fallback after dispatch to ${TO}. If ${WI_ID} is already done/verified, complete this check immediately — no action. Otherwise: if running but no output, re-prompt; if still queued, re-target or fail-fast. Do NOT create further check WorkItems for yourself." \
       '{
         type: $type,
         config: {type: "time", fireAt: $fireAt},

package/config/skills/orchestrator/delegate-task/execute.test.sh

@@ -152,7 +152,7 @@ OUT=$(CALL_LOG="$CALL_LOG" CREWLY_ROOT=/tmp/crewly-test \
 
 assert_log_contains "POST /triggers fires after dispatch by default" "POST /triggers"
 assert_output_contains "result reports fallbackTriggerId" "fallbackTriggerId" "$OUT"
-assert_output_contains "result reports fallbackMinutes=30 (default)" '"fallbackMinutes": 30' "$OUT"
+assert_output_contains "result reports fallbackMinutes=120 (default = §3.0 2× TL ETA)" '"fallbackMinutes": 120' "$OUT"
 
 # --- Test 4: fallback can be disabled via --fallback-minutes 0 ---
 > "$CALL_LOG"