@m13v/s4l 1.6.197-rc.10

package/skill/dm-outreach-twitter.sh

@@ -0,0 +1,265 @@
+#!/usr/bin/env bash
+# dm-outreach-twitter.sh — Outbound Twitter/X DM outreach.
+# Scans for DM candidates (users who engaged on our posts), then sends Twitter DMs
+# to continue the conversation. Inbound DM replies are handled separately
+# by engage-dm-replies-twitter.sh.
+# Called by launchd (com.m13v.social-dm-outreach-twitter) every 6 hours.
+
+set -euo pipefail
+
+# Cycle ID for cross-cycle cost accounting (see run-twitter-cycle.sh for the
+# same pattern). Stamps claude_sessions.cycle_id so get_run_cost.py --cycle-id
+# returns just this cycle's spend.
+BATCH_ID="${BATCH_ID:-dmtw-$(date +%Y%m%d-%H%M%S)}"
+export BATCH_ID
+export SA_CYCLE_ID="$BATCH_ID"
+
+# Bootstrap log paths early so the singleton-cleanup output below gets captured
+# in the same log file the rest of the run uses.
+LOG_DIR="$HOME/social-autoposter/skill/logs"
+mkdir -p "$LOG_DIR"
+LOG_FILE="$LOG_DIR/dm-outreach-twitter-$(date +%Y-%m-%d_%H%M%S).log"
+
+# Browser-profile lock first (shared with other twitter pipelines), then pipeline lock.
+source "$(dirname "$0")/lock.sh"
+# Harness-only browser bootstrap (twitter-agent path fully removed 2026-05-19).
+# Sets MCP_CONFIG_FILE, BROWSER_INSTRUCTIONS, exports TWITTER_CDP_URL=9555.
+source "$(dirname "$0")/lib/twitter-backend.sh"
+
+acquire_lock "twitter-browser" 3600
+ensure_twitter_browser_for_backend 2>&1 | tee -a "$LOG_FILE"
+acquire_lock "dm-outreach-twitter" 2700
+
+# Load secrets
+# shellcheck source=/dev/null
+[ -f "$HOME/social-autoposter/.env" ] && source "$HOME/social-autoposter/.env"
+
+REPO_DIR="$HOME/social-autoposter"
+SKILL_FILE="$REPO_DIR/SKILL.md"
+
+# 2026-06-02: removed the vestigial DATABASE_URL gate. This rail talks to the
+# central store exclusively through the S4L HTTP API (scan_dm_candidates.py,
+# dm_outreach_twitter_helper.py, log_run.py all use scripts/http_api.py). No
+# direct Postgres connection is opened here, matching dm-outreach-reddit.sh and
+# dm-outreach-linkedin.sh (migrated 2026-05-12).
+# (LOG_DIR/LOG_FILE bootstrapped at top of script.)
+
+log() { echo "[$(date +%H:%M:%S)] $*" | tee -a "$LOG_FILE"; }
+
+RUN_START=$(date +%s)
+log "=== Twitter DM Outreach Run: $(date) ==="
+
+# Scan for new DM candidates first (cheap Python, writes to dms table)
+log "Scanning for DM candidates (all platforms)..."
+(PYTHONUNBUFFERED=1 python3 "$REPO_DIR/scripts/scan_dm_candidates.py" 2>&1 || true) | tee -a "$LOG_FILE"
+
+DM_PENDING=$(python3 "$REPO_DIR/scripts/dm_outreach_twitter_helper.py" pending-count 2>/dev/null || echo "0")
+
+if [ "$DM_PENDING" -eq 0 ]; then
+    log "No pending Twitter DMs"
+    python3 "$REPO_DIR/scripts/log_run.py" --script "dm_outreach_twitter" --posted 0 --skipped 0 --failed 0 --cost 0 --elapsed $(( $(date +%s) - RUN_START ))
+    exit 0
+fi
+
+log "Twitter: $DM_PENDING DMs to send"
+
+# Prompt-feed JSON now comes from /api/v1/dms/outreach-queue (same
+# correlated other_engagement subquery, same 60-day window, same join
+# graph). Helper canonicalises platform=twitter → 'x' to match the dms
+# table's stored value.
+DM_DATA=$(python3 "$REPO_DIR/scripts/dm_outreach_twitter_helper.py" outreach-queue 2>/dev/null || echo "[]")
+
+# Per-project qualification context for ICP pre-check
+PROJECTS_QUALIFICATION=$(python3 -c "
+import json
+c = json.load(open('$REPO_DIR/config.json'))
+for p in c.get('projects', []):
+    q = p.get('qualification') or {}
+    if not q:
+        continue
+    print(f\"- {p['name']}:\")
+    if q.get('must_have'):
+        print(f\"    must_have: {' ; '.join(q['must_have'])}\")
+    if q.get('disqualify'):
+        print(f\"    disqualify: {' ; '.join(q['disqualify'])}\")
+" 2>/dev/null || echo "")
+
+export CLAUDE_SESSION_ID=$(uuidgen | tr 'A-Z' 'a-z')
+
+PROMPT_FILE=$(mktemp)
+cat > "$PROMPT_FILE" <<PROMPT_EOF
+You are the Social Autoposter Twitter/X DM outreach bot.
+
+Read $SKILL_FILE for content rules (tone, anti-AI detection, no em dashes).
+
+## Task: Send Twitter/X DMs to continue comment conversations
+
+These users engaged with our Twitter/X posts/comments. We already replied publicly. Now send a short, casual DM to continue the conversation.
+
+CRITICAL RULES:
+1. DMs must feel like a natural continuation of the comment discussion, NOT a cold outreach or sales pitch
+2. Reference the specific conversation topic, not generic "hey I saw your comment"
+3. Keep it short: 1-2 sentences max, like a text message
+4. No links in the first DM; earn the conversation first
+5. No em dashes. Write casually, like texting a coworker.
+
+## COMMITMENT GUARDRAILS (violating any of these is a critical failure)
+- **NEVER suggest, offer, or agree to calls, meetings, demos, or video chats.** Keep everything in the DM thread.
+- **NEVER agree to podcast appearances, X Spaces, interviews, or live events.**
+- **NEVER offer to move the conversation to another platform** (Telegram, Discord, email, etc.). Stay in the current DM thread.
+- **NEVER promise to share specific links, files, or resources you don't have right now.** If you don't have it in config.json projects, don't promise it.
+- **NEVER make time-bound commitments** ("this week", "tomorrow", "Thursday"). Don't commit the human to any schedule.
+- **NEVER say "I'm in [city]"** or share location/personal details not in config.json.
+- If someone asks for any of the above, respond naturally but deflect: keep the conversation going in the DM without making promises. Example: "honestly easier to hash it out here, what specifically are you trying to set up?"
+
+DM EXAMPLES (good):
+- "yo your point about token costs scaling with agent count hit home, we're dealing with the exact same thing. what's your setup look like?"
+- "that workaround you mentioned for the accessibility API crash is clever, did it hold up in production?"
+- "curious how you ended up going with that approach for the MCP server, we tried something similar"
+
+DM EXAMPLES (bad):
+- "Hey! I noticed your tweet. I'm building something you might find interesting..." (cold pitch)
+- "Great point! I'd love to connect and share what we're working on." (generic)
+- "Hi there, I saw your insightful tweet about AI agents..." (too formal)
+
+## Users to DM:
+$DM_DATA
+
+## Cross-thread engagement awareness
+Each row may include an \`other_engagement\` array: this user's other recent (60-day) interactions with our posts on the same platform. Each entry has thread_title, their_content snippet, our_reply_content snippet, depth (>1 = public follow-up to our reply in a thread), status, replied_at.
+
+Use it as context for the DM:
+- If the most recent other_engagement entry is on the SAME thread with depth>1 and replied_at < 6 hours ago, they're actively continuing the public conversation. Prefer a lighter-touch DM, or open with an acknowledgment of the ongoing thread instead of introducing a new angle.
+- If they've engaged on multiple other threads, it signals genuine interest. The DM can be slightly more direct without feeling cold.
+- Do NOT quote their other comments back at them or enumerate their history. It's context, not content.
+
+## Per-project ICP criteria (used for the pre-check step, NOT to skip sending):
+$PROJECTS_QUALIFICATION
+
+## Pre-send profile fetch + ICP pre-check (MANDATORY per DM, no filter)
+
+For each DM row, BEFORE you compose or send, do this in order:
+
+1. Look at the row's \`target_project\`. If it's NULL, set icp_precheck=unknown with notes="no_target_project" and proceed to step 4 — but still try to capture profile basics.
+
+2. Fetch the prospect's X/Twitter profile using the browser tools from the BROWSER BACKEND block above:
+   - Navigate to https://x.com/THEIR_AUTHOR (strip any leading @).
+   - Snapshot the page. Extract: display name, handle, bio text, follower count, pinned/top-of-feed recent tweet topic summary.
+   - If the profile is suspended, protected, or empty, capture what you can and note "profile_limited" or "profile_inaccessible".
+
+3. Persist the profile fields:
+   \`\`\`bash
+   python3 $REPO_DIR/scripts/fetch_prospect_profile.py upsert \\
+       --platform twitter --author "THEIR_AUTHOR" \\
+       --profile-url "https://x.com/THEIR_AUTHOR" \\
+       --display-name "DISPLAY_NAME" \\
+       --headline "SHORT_BIO_FIRST_LINE" \\
+       --bio "FULL_BIO_TEXT" \\
+       --follower-count N \\
+       --recent-activity "SHORT_RECENT_TWEETS_SUMMARY" \\
+       --notes "ANY_SIGNAL_WORTH_REMEMBERING" \\
+       --link-dm DM_ID
+   \`\`\`
+   Omit any flag whose value is empty or unknown. \`--link-dm\` also wires dms.prospect_id.
+
+4. Evaluate ICP match against EVERY project listed in "Per-project ICP criteria" above (not only target_project). For each project compare the profile + their_content + comment_context against its must_have (satisfy at least one) and disqualify (trigger ANY = fail), and pick one label: icp_match, icp_miss, disqualified, or unknown. Upsert one entry per project:
+   \`\`\`bash
+   python3 $REPO_DIR/scripts/dm_conversation.py set-icp-precheck \\
+       --dm-id DM_ID --project PROJECT_NAME --label LABEL --notes "SHORT_RATIONALE"
+   \`\`\`
+   Run this once per project from the list. Each call upserts one entry in dms.icp_matches (JSONB array) keyed by project.
+
+5. If ANY entry in icp_matches has label=disqualified, skip the send: run \`python3 scripts/dm_conversation.py mark-skipped --dm-id DM_ID --reason "disqualified: PROJECT - SHORT_NOTES"\` and move on. \`icp_miss\` alone does NOT gate; send when every project scored miss. Only explicit \`disqualified\` blocks the opener.
+
+## How to send Twitter/X DMs (use the browser tools from the BROWSER BACKEND block):
+1. Navigate to https://x.com/messages
+2. **ENCRYPTED DM PASSCODE**: Twitter may show an "Enter your passcode" or "encrypted_dm_passcode_required" dialog before you can access DMs. If you see this dialog:
+   a. Find the passcode input field in the snapshot
+   b. Type the passcode: $TWITTER_DM_PASSCODE
+   c. Click "Confirm" or press Enter
+   d. Wait for the DM inbox to load
+   The passcode is loaded from .env as TWITTER_DM_PASSCODE.
+3. Start new message to THEIR_AUTHOR
+4. Type and send the message.
+
+## After each DM:
+
+Inspect the send_dm tool's return value. There are exactly three outcomes:
+
+(A) ok=true AND verified=true  ->  success, mark sent:
+  CLAUDE_SESSION_ID=$CLAUDE_SESSION_ID python3 $REPO_DIR/scripts/dm_send_log.py \\
+      --dm-id DM_ID --message "DM_TEXT" --verified
+
+  Do NOT issue a raw "UPDATE dms SET status='sent'" psql command. The
+  dm_send_log.py script is the only path that may flip status to 'sent';
+  it requires --verified, and refuses without it. This is intentional:
+  prior phantom-DM bugs (~700 rows in 4/2026) came from prose-driven
+  status flips that ignored the verification result.
+
+  After the DM lands, capture the current page URL by evaluating window.location.href in the browser (use the JS-eval tool from the BROWSER BACKEND block) and stamp it onto the DM row so the dashboard's "open chat" button works:
+  python3 $REPO_DIR/scripts/dm_conversation.py set-url --dm-id DM_ID --url "CHAT_URL"
+  The validator only accepts /i/chat/<id> or /messages/<id>; if the URL is something else (you got bounced to a profile or inbox), skip this step.
+
+(B) ok=false OR verified=false  ->  send did not land, mark error via /api/v1/dms/DM_ID PATCH (DO NOT shell out to psql):
+  python3 $REPO_DIR/scripts/dm_db_update.py --dm-id DM_ID --status error --skip-reason send_unverified --claude-session-id "$CLAUDE_SESSION_ID"
+
+(C) Rate limit, account blocked, or any other thrown exception:
+  python3 $REPO_DIR/scripts/dm_db_update.py --dm-id DM_ID --status error --skip-reason REASON --claude-session-id "$CLAUDE_SESSION_ID"
+
+DMs disabled (recipient setting, not a send failure):
+  python3 $REPO_DIR/scripts/dm_db_update.py --dm-id DM_ID --status skipped --skip-reason chat_disabled --claude-session-id "$CLAUDE_SESSION_ID"
+
+CRITICAL: ALL browser calls MUST use the tools listed in the BROWSER BACKEND block above (the Twitter-dedicated MCP for this run). NEVER use generic mcp__playwright-extension__*, mcp__isolated-browser__*, or mcp__macos-use__* tools. If a browser tool call is blocked or times out, wait 30 seconds and retry (up to 3 times). Do NOT fall back to any other browser tool.
+
+## CRITICAL FAILURE MODE: Twitter browser tools not registered
+If at the START of this run you cannot see ANY of the browser tools listed in the BROWSER BACKEND block above, OR every browser call fails with "MCP server not connected" / "no such tool" / similar, this is a transient infrastructure failure (Chrome profile collision, wedged MCP wrapper, lock acquired but profile still held by another process). It is NOT an error condition for the prospects in the queue.
+
+Do EXACTLY this:
+1. Make NO database changes. Do NOT mark any row as 'error', 'skipped', or anything else.
+2. Print a single line to stdout: \`MCP_UNAVAILABLE: twitter browser tools not registered; aborting with rows left at status=pending\`
+3. Exit cleanly. The launchd schedule will retry on the next cycle, by which point the wedged profile holder will likely have timed out.
+
+Burning rows with skip_reason='twitter_agent_mcp_unavailable: ...' is a regression that on 2026-05-12 nuked 7 warm leads (efemjoba, gpuops, josesaezmerino, AIDailyGems, alkimiadev, RobertDMellish, kunaljeweller). The d.id IS NULL filter in scan_dm_candidates.py then permanently blocked them from re-discovery. Do not do this.
+PROMPT_EOF
+
+gtimeout 2700 "$REPO_DIR/scripts/run_claude.sh" "dm-outreach-twitter" --strict-mcp-config --mcp-config "$MCP_CONFIG_FILE" --output-format stream-json --verbose -p "${BROWSER_INSTRUCTIONS}
+
+$(cat "$PROMPT_FILE")" 2>&1 | tee -a "$LOG_FILE" || log "WARNING: Twitter DM outreach claude exited with code $?"
+rm -f "$PROMPT_FILE"
+
+# Belt-and-suspenders: if Claude (despite the prompt instructions added 2026-05-13)
+# marked any row with skip_reason='twitter_agent_mcp_unavailable: ...' or
+# 'mcp_unavailable' or similar transient-MCP signals THIS run, revert it back
+# to status='pending' so the next cycle can retry. Scoped to this run via
+# claude_session_id to avoid clobbering legitimate older rows.
+#
+# The scan_dm_candidates.py discover query uses `LEFT JOIN dms ... WHERE d.id IS NULL`
+# so once a reply has any dms row (even status='error'), it never re-appears as a
+# candidate. That's how the 2026-05-12 incident permanently lost 7 warm leads.
+# This revert closes the gap by ensuring transient-MCP failures don't lock rows
+# into a status=error state that the scanner can't see past.
+# MCP-failure recovery sweep now lives at /api/v1/dms/recover-mcp-failures
+# (same UPDATE WHERE filter, same RETURNING shape). Helper prints the
+# recovered_count integer so this $() capture is byte-equivalent.
+RECOVERED=$(python3 "$REPO_DIR/scripts/dm_outreach_twitter_helper.py" \
+    recover-mcp --session-id "$CLAUDE_SESSION_ID" 2>/dev/null || echo "0")
+if [ "$RECOVERED" -gt 0 ]; then
+    log "Reverted $RECOVERED row(s) from status='error' (transient MCP failure) back to pending"
+fi
+
+# Final summary counts: one HTTP roundtrip via /api/v1/dms/counts vs the
+# two psql one-liners. Helper prints "<sent> <pending>" space-separated.
+_DM_SUMMARY=$(python3 "$REPO_DIR/scripts/dm_outreach_twitter_helper.py" summary 2>/dev/null || echo "0 0")
+SENT=$(echo "$_DM_SUMMARY" | awk '{print $1}')
+STILL_PENDING=$(echo "$_DM_SUMMARY" | awk '{print $2}')
+: "${SENT:=0}"
+: "${STILL_PENDING:=0}"
+log "Twitter DM outreach summary: sent (all-time)=$SENT, still_pending=$STILL_PENDING"
+
+RUN_ELAPSED=$(( $(date +%s) - RUN_START ))
+_COST=$(python3 "$REPO_DIR/scripts/get_run_cost.py" --since "$RUN_START" --scripts "dm-outreach-twitter" 2>/dev/null || echo "0.0000")
+python3 "$REPO_DIR/scripts/log_run.py" --script "dm_outreach_twitter" --posted 0 --skipped 0 --failed 0 --cost "$_COST" --elapsed "$RUN_ELAPSED"
+
+find "$LOG_DIR" -name "dm-outreach-twitter-*.log" -mtime +7 -delete 2>/dev/null || true
+
+log "=== Twitter DM outreach complete: $(date) ==="

package/skill/engage-dm-replies-linkedin.sh

@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+# engage-dm-replies-linkedin.sh — LinkedIn DM replies only
+
+exec "$(dirname "$0")/engage-dm-replies.sh" --platform linkedin

package/skill/engage-dm-replies-reddit.sh

@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+# engage-dm-replies-reddit.sh — Reddit DM replies only
+
+exec "$(dirname "$0")/engage-dm-replies.sh" --platform reddit

package/skill/engage-dm-replies-twitter.sh

@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+# engage-dm-replies-twitter.sh — Twitter DM replies only
+
+exec "$(dirname "$0")/engage-dm-replies.sh" --platform twitter