diff --git a/formulas/run-predictor.toml b/formulas/run-predictor.toml
new file mode 100644
index 0000000..d16e883
--- /dev/null
+++ b/formulas/run-predictor.toml
@@ -0,0 +1,194 @@
+# formulas/run-predictor.toml — Predictor formula (disinto-specific signals)
+#
+# Executed by predictor/predictor-run.sh via cron — no action issues.
+# predictor-run.sh creates a tmux session with Claude (sonnet) and injects
+# this formula as context. Claude executes all steps autonomously.
+#
+# Steps: preflight → collect-signals → analyze-and-predict
+#
+# Disinto-specific signal sources:
+#   - CI pipeline trends (Woodpecker)
+#   - Stale issues (open issues with no recent activity)
+#   - Agent health (tmux sessions, recent logs)
+#   - Resource patterns (RAM, disk, load, containers)
+
+name        = "run-predictor"
+description = "Evidence-based prediction: CI trends, stale issues, agent health, resource patterns"
+version     = 1
+model       = "sonnet"
+
+[context]
+files = ["AGENTS.md", "RESOURCES.md"]
+
+[[steps]]
+id    = "preflight"
+title = "Pull latest code and gather environment"
+description = """
+Set up the working environment for this prediction run.
+
+1. Change to the project repository:
+     cd "$PROJECT_REPO_ROOT"
+
+2. Pull the latest code:
+     git fetch origin "$PRIMARY_BRANCH" --quiet
+     git checkout "$PRIMARY_BRANCH" --quiet
+     git pull --ff-only origin "$PRIMARY_BRANCH" --quiet
+"""
+
+[[steps]]
+id    = "collect-signals"
+title = "Collect disinto-specific signals"
+description = """
+Gather raw signal data for pattern analysis. Collect each signal category
+and store the results for the analysis step.
+
+### 1. CI pipeline trends (Woodpecker)
+
+Fetch recent builds from Woodpecker CI:
+  curl -sf -H "Authorization: Bearer $WOODPECKER_TOKEN" \
+    "${WOODPECKER_URL}/api/repos/${WOODPECKER_REPO_ID}/pipelines?page=1&perPage=20"
+
+Look for:
+- Build failure rate over last 20 builds
+- Repeated failures on the same step
+- Builds stuck in running/pending state
+- Time since last successful build
+
+If WOODPECKER_TOKEN or WOODPECKER_URL are not set, skip CI signals and note
+"CI signals unavailable — WOODPECKER_TOKEN not configured".
+
+### 2. Stale issues
+
+Fetch all open issues:
+  curl -sf -H "Authorization: token $CODEBERG_TOKEN" \
+    "$CODEBERG_API/issues?state=open&type=issues&limit=50&sort=updated&direction=asc"
+
+Identify:
+- Issues with no update in 14+ days (stale)
+- Issues with no update in 30+ days (very stale)
+- Issues labeled 'action' or 'backlog' that are stale (work not progressing)
+- Blocked issues where the blocker may have been resolved
+
+### 3. Agent health
+
+Check active tmux sessions:
+  tmux list-sessions 2>/dev/null || echo "no sessions"
+
+Check recent agent logs (last 24h of activity):
+  for log in supervisor/supervisor.log planner/planner.log planner/prediction.log \
+             gardener/gardener.log dev/dev.log review/review.log; do
+    if [ -f "$PROJECT_REPO_ROOT/$log" ]; then
+      echo "=== $log (last 20 lines) ==="
+      tail -20 "$PROJECT_REPO_ROOT/$log"
+    fi
+  done
+
+Look for:
+- Agents that haven't run recently (missing log entries in last 24h)
+- Repeated errors or failures in logs
+- Sessions stuck or crashed (tmux sessions present but no recent activity)
+- Lock files that may be stale: /tmp/*-poll.lock, /tmp/*-run.lock
+
+### 4. Resource patterns
+
+Collect current resource state:
+  free -m                          # RAM
+  df -h /                          # Disk
+  cat /proc/loadavg                # Load average
+  docker ps --format '{{.Names}} {{.Status}}' 2>/dev/null || true  # Containers
+
+Look for:
+- Available RAM < 2000MB (agents will skip runs)
+- Disk usage > 80% (approaching danger zone)
+- Load average > 3.0 (box overloaded)
+- Containers in unhealthy or restarting state
+
+### 5. Already-open predictions (deduplication)
+
+Fetch existing open predictions to avoid duplicates:
+  curl -sf -H "Authorization: token $CODEBERG_TOKEN" \
+    "$CODEBERG_API/issues?state=open&type=issues&labels=prediction%2Funreviewed&limit=50"
+
+Also check prediction/backlog (watched but not yet actioned):
+  curl -sf -H "Authorization: token $CODEBERG_TOKEN" \
+    "$CODEBERG_API/issues?state=open&type=issues&labels=prediction%2Fbacklog&limit=50"
+
+Record their titles so you can avoid duplicating them.
+"""
+needs = ["preflight"]
+
+[[steps]]
+id    = "analyze-and-predict"
+title = "Analyze signals and file prediction issues"
+description = """
+Analyze the collected signals for patterns and file up to 5 prediction issues.
+
+## What to look for
+
+**CI regression** — Build failure rate increasing or repeated failures:
+- Failure rate > 30% over last 20 builds → high confidence
+- Same step failing 3+ times in a row → high confidence
+- No successful build in 24+ hours → medium confidence
+
+**Stale work** — Issues not progressing:
+- Action issues stale 7+ days → the action agent may be stuck
+- Backlog issues stale 14+ days → work not being picked up
+- Blocked issues whose blockers are now closed → can be unblocked
+
+**Agent health** — Agents not running or failing:
+- Agent log with no entries in 24+ hours → agent may be down
+- Repeated errors in agent logs → systemic problem
+- Stale lock files (process not running but lock exists)
+
+**Resource pressure** — System approaching limits:
+- RAM < 2000MB → agents will start skipping runs
+- Disk > 80% → approaching critical threshold
+- Load sustained > 3.0 → box is overloaded, queued work backing up
+
+**Opportunity** — Good conditions for expensive work:
+- Box idle (RAM > 3000MB, load < 1.0, few active sessions) → good time
+  for expensive operations if any are pending
+
+## Filing predictions
+
+For each prediction, create a Codeberg issue with the `prediction/unreviewed` label.
+
+1. Look up the label ID:
+     curl -sf -H "Authorization: token $CODEBERG_TOKEN" \
+       "$CODEBERG_API/labels" | jq '.[] | select(.name == "prediction/unreviewed") | .id'
+
+2. For each prediction, create an issue:
+     curl -sf -X POST -H "Authorization: token $CODEBERG_TOKEN" \
+       -H "Content-Type: application/json" \
+       "$CODEBERG_API/issues" \
+       -d '{"title":"<title>","body":"<body>","labels":[<label_id>]}'
+
+   Body format:
+     <2-4 sentence description of what was observed, why it matters,
+      what the planner should consider>
+
+     ---
+     **Signal source:** <which signal triggered this>
+     **Confidence:** <high|medium|low>
+     **Suggested action:** <concrete next step for the planner>
+
+3. Send a Matrix notification for each prediction created (optional):
+     Use matrix_send if available, or skip if MATRIX_TOKEN is not set.
+
+## Rules
+- Max 5 predictions total
+- Do NOT predict feature work — only infrastructure/health/metric observations
+- Do NOT duplicate existing open predictions (checked in collect-signals)
+- Be specific: name the metric, the value, the threshold
+- Prefer high-confidence predictions backed by concrete data
+- If no meaningful patterns found, file zero issues — that is a valid outcome
+
+## Completion
+
+When done (whether predictions were filed or not):
+  echo 'PHASE:done' > '$PHASE_FILE'
+
+On unrecoverable error:
+  printf 'PHASE:failed\nReason: %s\n' 'describe error' > '$PHASE_FILE'
+"""
+needs = ["collect-signals"]
diff --git a/predictor/predictor-run.sh b/predictor/predictor-run.sh
new file mode 100755
index 0000000..381a3f9
--- /dev/null
+++ b/predictor/predictor-run.sh
@@ -0,0 +1,115 @@
+#!/usr/bin/env bash
+# =============================================================================
+# predictor-run.sh — Cron wrapper: predictor execution via Claude + formula
+#
+# Runs daily (or on-demand). Guards against concurrent runs and low memory.
+# Creates a tmux session with Claude (sonnet) reading formulas/run-predictor.toml.
+# Files prediction/unreviewed issues for the planner to triage.
+#
+# Usage:
+#   predictor-run.sh [projects/disinto.toml]   # project config (default: disinto)
+#
+# Cron: 0 6 * * * cd /path/to/dark-factory && bash predictor/predictor-run.sh
+# =============================================================================
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+FACTORY_ROOT="$(dirname "$SCRIPT_DIR")"
+
+# Accept project config from argument; default to disinto
+export PROJECT_TOML="${1:-$FACTORY_ROOT/projects/disinto.toml}"
+# shellcheck source=../lib/env.sh
+source "$FACTORY_ROOT/lib/env.sh"
+# shellcheck source=../lib/agent-session.sh
+source "$FACTORY_ROOT/lib/agent-session.sh"
+# shellcheck source=../lib/formula-session.sh
+source "$FACTORY_ROOT/lib/formula-session.sh"
+
+LOG_FILE="$SCRIPT_DIR/predictor.log"
+SESSION_NAME="predictor-${PROJECT_NAME}"
+PHASE_FILE="/tmp/predictor-session-${PROJECT_NAME}.phase"
+
+# shellcheck disable=SC2034  # read by monitor_phase_loop in lib/agent-session.sh
+PHASE_POLL_INTERVAL=15
+
+log() { echo "[$(date -u +%Y-%m-%dT%H:%M:%S)Z] $*" >> "$LOG_FILE"; }
+
+# ── Guards ────────────────────────────────────────────────────────────────
+acquire_cron_lock "/tmp/predictor-run.lock"
+check_memory 2000
+
+log "--- Predictor run start ---"
+
+# ── Load formula + context ───────────────────────────────────────────────
+load_formula "$FACTORY_ROOT/formulas/run-predictor.toml"
+build_context_block AGENTS.md RESOURCES.md
+
+# ── Build prompt ─────────────────────────────────────────────────────────
+PROMPT="You are the prediction agent (goblin) for ${CODEBERG_REPO}. Work through the formula below. You MUST write PHASE:done to '${PHASE_FILE}' when finished — the orchestrator will time you out if you return to the prompt without signalling.
+
+Your role: spot patterns in infrastructure signals and file them as prediction issues.
+The planner (adult) will triage every prediction before acting.
+You MUST NOT emit feature work or implementation issues — only predictions
+about CI health, issue staleness, agent status, and system conditions.
+
+## Project context
+${CONTEXT_BLOCK}
+
+## Formula
+${FORMULA_CONTENT}
+
+## Codeberg API reference
+Base URL: ${CODEBERG_API}
+Auth header: -H \"Authorization: token \$CODEBERG_TOKEN\"
+  Read issue:  curl -sf -H \"Authorization: token \$CODEBERG_TOKEN\" '${CODEBERG_API}/issues/{number}' | jq '.body'
+  Create issue: curl -sf -X POST -H \"Authorization: token \$CODEBERG_TOKEN\" -H 'Content-Type: application/json' '${CODEBERG_API}/issues' -d '{\"title\":\"...\",\"body\":\"...\",\"labels\":[LABEL_ID]}'
+  List labels: curl -sf -H \"Authorization: token \$CODEBERG_TOKEN\" '${CODEBERG_API}/labels'
+NEVER echo or include the actual token value in output — always reference \$CODEBERG_TOKEN.
+
+## Environment
+FACTORY_ROOT=${FACTORY_ROOT}
+PROJECT_REPO_ROOT=${PROJECT_REPO_ROOT}
+PRIMARY_BRANCH=${PRIMARY_BRANCH}
+PHASE_FILE=${PHASE_FILE}
+
+## Phase protocol (REQUIRED)
+When all work is done:
+  echo 'PHASE:done' > '${PHASE_FILE}'
+On unrecoverable error:
+  printf 'PHASE:failed\nReason: %s\n' 'describe error' > '${PHASE_FILE}'"
+
+# ── Create tmux session ─────────────────────────────────────────────────
+export CLAUDE_MODEL="sonnet"
+if ! start_formula_session "$SESSION_NAME" "$PROJECT_REPO_ROOT" "$PHASE_FILE"; then
+  exit 1
+fi
+
+agent_inject_into_session "$SESSION_NAME" "$PROMPT"
+log "Prompt sent to tmux session"
+matrix_send "predictor" "Predictor session started for ${CODEBERG_REPO}" 2>/dev/null || true
+
+# ── Phase monitoring loop ────────────────────────────────────────────────
+log "Monitoring phase file: ${PHASE_FILE}"
+_FORMULA_CRASH_COUNT=0
+
+monitor_phase_loop "$PHASE_FILE" 7200 "formula_phase_callback"
+
+FINAL_PHASE=$(read_phase "$PHASE_FILE")
+log "Final phase: ${FINAL_PHASE:-none}"
+
+if [ "$FINAL_PHASE" != "PHASE:done" ]; then
+  case "${_MONITOR_LOOP_EXIT:-}" in
+    idle_prompt)
+      log "predictor: Claude returned to prompt without writing phase signal"
+      ;;
+    idle_timeout)
+      log "predictor: timed out after 2h with no phase signal"
+      ;;
+    *)
+      log "predictor finished without PHASE:done (phase: ${FINAL_PHASE:-none}, exit: ${_MONITOR_LOOP_EXIT:-})"
+      ;;
+  esac
+fi
+
+matrix_send "predictor" "Predictor session finished (${FINAL_PHASE:-no phase})" 2>/dev/null || true
+log "--- Predictor run done ---"