social-autoposter 1.6.1 → 1.6.3

package/bin/server.js

@@ -49,6 +49,13 @@ const JOBS = [
   // Post Threads row (original threads/posts)
   { label: 'com.m13v.social-reddit-threads', name: 'Reddit Threads', type: 'Post Threads', platform: 'Reddit', script: 'run-reddit-threads.sh', logPrefix: 'run-reddit-threads-', plist: 'com.m13v.social-reddit-threads.plist' },
   { label: 'com.m13v.social-twitter-threads', name: 'Twitter Threads', type: 'Post Threads', platform: 'Twitter', script: 'run-twitter-threads.sh', logPrefix: 'run-twitter-threads-', plist: 'com.m13v.social-twitter-threads.plist' },
+  // Instagram per-account daily posters (5×/day each, FORCE_ACCOUNT pinned).
+  { label: 'com.m13v.social-instagram-daily-matt_diak', name: 'IG Daily (matt_diak)', type: 'Post Threads', platform: 'Instagram', script: 'run-instagram-daily.sh', logPrefix: 'instagram-daily-', plist: 'com.m13v.social-instagram-daily-matt_diak.plist' },
+  { label: 'com.m13v.social-instagram-daily-matthewheartful', name: 'IG Daily (matthewheartful)', type: 'Post Threads', platform: 'Instagram', script: 'run-instagram-daily.sh', logPrefix: 'instagram-daily-', plist: 'com.m13v.social-instagram-daily-matthewheartful.plist' },
+  // Instagram per-account render (upstream of daily-posters; produces the
+  // mp4 + caption draft that the daily-poster then uploads).
+  { label: 'com.m13v.social-instagram-render-matt_diak', name: 'IG Render (matt_diak)', type: 'Other', platform: 'Instagram', script: 'run-instagram-render.sh', logPrefix: 'instagram-render-', plist: 'com.m13v.social-instagram-render-matt_diak.plist' },
+  { label: 'com.m13v.social-instagram-render-matthewheartful', name: 'IG Render (matthewheartful)', type: 'Other', platform: 'Instagram', script: 'run-instagram-render.sh', logPrefix: 'instagram-render-', plist: 'com.m13v.social-instagram-render-matthewheartful.plist' },
   // Post Comments row (replies/comments on others' content)
   { label: 'com.m13v.social-reddit-search', name: 'Reddit', type: 'Post Comments', platform: 'Reddit', script: 'run-reddit-search.sh', logPrefix: 'run-reddit-search-', plist: 'com.m13v.social-reddit-search.plist' },
   { label: 'com.m13v.social-twitter-cycle', name: 'Twitter', type: 'Post Comments', platform: 'Twitter', script: 'run-twitter-cycle.sh', logPrefix: 'twitter-cycle-', plist: 'com.m13v.social-twitter-cycle.plist' },
@@ -64,6 +71,8 @@ const JOBS = [
   { label: 'com.m13v.social-engage-reddit', name: 'Engage Reddit', type: 'Engage', platform: 'Reddit', script: 'engage-reddit.sh', logPrefix: 'engage-reddit-', plist: 'com.m13v.social-engage-reddit.plist' },
   { label: 'com.m13v.social-scan-moltbook-replies', name: 'MoltBook Scan', type: 'Other', platform: 'MoltBook', script: 'run-scan-moltbook-replies.sh', logPrefix: 'run-scan-moltbook-replies-', plist: 'com.m13v.social-scan-moltbook-replies.plist' },
   { label: 'com.m13v.social-scan-twitter-followups', name: 'Twitter Thread Follow-ups', type: 'Other', platform: 'Twitter', script: 'scan-twitter-followups.sh', logPrefix: 'scan-twitter-followups-', plist: 'com.m13v.social-scan-twitter-followups.plist' },
+  { label: 'com.m13v.social-scan-instagram-replies', name: 'Instagram Scan', type: 'Other', platform: 'Instagram', script: 'scan-instagram-replies.sh', logPrefix: 'scan-instagram-replies-', plist: 'com.m13v.social-scan-instagram-replies.plist' },
+  { label: 'com.m13v.social-refresh-instagram-tokens', name: 'IG Token Refresh', type: 'Other', platform: 'Instagram', script: 'refresh-instagram-tokens.sh', logPrefix: 'refresh-instagram-tokens-', plist: 'com.m13v.social-refresh-instagram-tokens.plist' },
   // DM Outreach row (initiate DMs to engaged users)
   { label: 'com.m13v.social-dm-outreach-reddit', name: 'DM Outreach Reddit', type: 'DM Outreach', platform: 'Reddit', script: 'dm-outreach-reddit.sh', logPrefix: 'dm-outreach-reddit-', plist: 'com.m13v.social-dm-outreach-reddit.plist' },
   { label: 'com.m13v.social-dm-outreach-twitter', name: 'DM Outreach Twitter', type: 'DM Outreach', platform: 'Twitter', script: 'dm-outreach-twitter.sh', logPrefix: 'dm-outreach-twitter-', plist: 'com.m13v.social-dm-outreach-twitter.plist' },
@@ -130,6 +139,10 @@ const REQUIRED_LOCKS = {
   'link-edit-github.sh':            ['link-edit-github'],
   'stats-reddit.sh':                ['reddit-browser'],
   'stats-instagram.sh':             ['instagram-poster'],
+  'scan-instagram-replies.sh':      ['instagram-poster'],
+  'refresh-instagram-tokens.sh':    ['instagram-poster'],
+  'run-instagram-daily.sh':         ['instagram-poster'],
+  'run-instagram-render.sh':        ['instagram-render'],
   'audit-reddit.sh':                ['reddit-browser', 'audit-reddit'],
   'audit-twitter.sh':               ['twitter-browser', 'audit-twitter'],
   'audit-linkedin.sh':              ['linkedin-browser', 'audit-linkedin'],
@@ -5234,7 +5247,7 @@ async function handleApi(req, res) {
     const url = new URL(req.url, 'http://localhost');
     const windowHours = Math.max(1, Math.min(720, parseInt(url.searchParams.get('hours') || '24', 10) || 24));
     const rawProject = (url.searchParams.get('project') || '').trim();
-    const ALLOWED_COST_PLATFORMS = new Set(['reddit', 'twitter', 'linkedin', 'moltbook', 'github', 'seo', 'email']);
+    const ALLOWED_COST_PLATFORMS = new Set(['reddit', 'twitter', 'linkedin', 'moltbook', 'github', 'seo', 'email', 'instagram']);
     let rawPlat = String(url.searchParams.get('platform') || '').toLowerCase().trim();
     if (rawPlat === 'x') rawPlat = 'twitter';
     const plat = ALLOWED_COST_PLATFORMS.has(rawPlat) ? rawPlat : '';
@@ -5376,7 +5389,7 @@ async function handleApi(req, res) {
     const project = (rawProject === '' || rawProject.toLowerCase() === 'all') ? '' : rawProject;
     const projectOk = project === '' || /^[A-Za-z0-9_\- ]{1,64}$/.test(project);
     if (!projectOk) return json(res, { error: 'invalid project' }, 400);
-    const ALLOWED_COST_PLATFORMS = new Set(['reddit', 'twitter', 'linkedin', 'moltbook', 'github', 'seo', 'email']);
+    const ALLOWED_COST_PLATFORMS = new Set(['reddit', 'twitter', 'linkedin', 'moltbook', 'github', 'seo', 'email', 'instagram']);
     let rawPlat = String(url.searchParams.get('platform') || '').toLowerCase().trim();
     if (rawPlat === 'x') rawPlat = 'twitter';
     const plat = ALLOWED_COST_PLATFORMS.has(rawPlat) ? rawPlat : '';
@@ -6751,7 +6764,7 @@ async function handleApi(req, res) {
     const configuredProjects = Array.isArray(config.projects) ? config.projects : [];
     const weighted = configuredProjects.filter(p => (p.weight || 0) > 0);
     const totalWeight = weighted.reduce((a, p) => a + (p.weight || 0), 0) || 1;
-    const platforms = ['reddit', 'twitter', 'linkedin', 'moltbook', 'github'];
+    const platforms = ['reddit', 'twitter', 'linkedin', 'moltbook', 'github', 'instagram'];
     // Per-platform eligibility: a project is eligible to be picked for a
     // platform only if it has the data that platform's picker needs. Mirrors
     // scripts/pick_project.py and scripts/pick_thread_target.py. Projects
@@ -8093,6 +8106,7 @@ const HTML = `<!DOCTYPE html>
           <th>LinkedIn</th>
           <th>MoltBook</th>
           <th>GitHub</th>
+          <th>Instagram</th>
         </tr>
       </thead>
       <tbody id="matrix-body"></tbody>
@@ -10574,8 +10588,8 @@ const EVENT_DESCRIPTIONS = {
   page_expired: 'SEO page deleted by the daily expire pipeline because it had zero clicks in the last 30 days. The on-disk source file was removed; Next.js now returns 404 for the URL. Logged for audit/revert in seo_expired_pages.',
   resurrected: 'Previously archived or unavailable item brought back into rotation (e.g., a removed post restored after reappearing).',
 };
-const ACTIVITY_PLATFORMS = ['reddit', 'twitter', 'linkedin', 'moltbook', 'github', 'seo'];
-const ACTIVITY_PLATFORM_LABELS = { reddit: 'Reddit', twitter: 'Twitter / X', linkedin: 'LinkedIn', moltbook: 'Moltbook', github: 'GitHub', seo: 'SEO' };
+const ACTIVITY_PLATFORMS = ['reddit', 'twitter', 'linkedin', 'moltbook', 'github', 'seo', 'instagram'];
+const ACTIVITY_PLATFORM_LABELS = { reddit: 'Reddit', twitter: 'Twitter / X', linkedin: 'LinkedIn', moltbook: 'Moltbook', github: 'GitHub', seo: 'SEO', instagram: 'Instagram' };
 const PROJECT_LABELS = { tenxats: '10xats' };
 const ACTIVITY_PROJECT_NONE = '(none)';
 const ACTIVITY_CAMPAIGN_ORGANIC = '(organic)';
@@ -16515,10 +16529,10 @@ async function loadDeployHealth() {
 // hours by platform against config.json weight targets. Each platform cell
 // shows the count plus that project's share of the platform's posts in
 // brackets, so operators can spot imbalance without a separate deficit field.
-const PROJECT_STATUS_PLATFORMS = ['reddit', 'twitter', 'linkedin', 'moltbook', 'github'];
+const PROJECT_STATUS_PLATFORMS = ['reddit', 'twitter', 'linkedin', 'moltbook', 'github', 'instagram'];
 const PROJECT_STATUS_PLATFORM_LABELS = {
   reddit: 'Reddit', twitter: 'Twitter', linkedin: 'LinkedIn',
-  moltbook: 'MoltBook', github: 'GitHub',
+  moltbook: 'MoltBook', github: 'GitHub', instagram: 'Instagram',
 };
 let _projectStatusLoading = false;
 let _projectStatusData = null;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "social-autoposter",
-  "version": "1.6.1",
+  "version": "1.6.3",
   "description": "Automated social posting pipeline for Reddit, X/Twitter, LinkedIn, and Moltbook. Install as a Claude Code agent skill.",
   "bin": {
     "social-autoposter": "bin/cli.js"

package/scripts/refresh_instagram_tokens.py

@@ -0,0 +1,280 @@
+#!/usr/bin/env python3
+"""Refresh Instagram Graph API long-lived tokens before they expire.
+
+Instagram long-lived user tokens are valid for ~60 days. Each call to the
+refresh_access_token endpoint extends the lifetime by another 60 days. The
+token must be at least 24 hours old to be refreshable, and Meta recommends
+refreshing well before expiry (we use a 14-day buffer).
+
+This script:
+  1. Iterates over every account in config.json -> instagram.accounts[].
+  2. Reads the current token + expiry from ~/instagram-graph-api/.env via the
+     ig_long_token_env / derived IG_TOKEN_EXPIRES_<suffix> key.
+  3. If the token expires within REFRESH_BUFFER_DAYS, calls the Graph API
+     refresh_access_token endpoint and rewrites the .env file in place
+     (atomic: write to tempfile then os.replace).
+  4. Prints a machine-readable SUMMARY line for the wrapper to log via
+     scripts/log_run.py.
+
+The .env file is the SINGLE source of truth — update_instagram_stats.py and
+scan_instagram_comments.py both read it on every invocation, so a refreshed
+token is picked up by the next pipeline run with no daemon-restart needed.
+
+Usage:
+    python3 scripts/refresh_instagram_tokens.py [--quiet] [--force] [--dry-run]
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import os
+import sys
+import tempfile
+import time
+import urllib.error
+import urllib.parse
+import urllib.request
+from datetime import datetime, timezone
+from pathlib import Path
+
+IG_ENV_PATH = Path.home() / "instagram-graph-api" / ".env"
+GRAPH = "https://graph.instagram.com"
+SA_CONFIG = Path(__file__).resolve().parent.parent / "config.json"
+
+# Refresh tokens that expire within this many days. 14 days gives us 2 weeks
+# of headroom for cron failures, network outages, or attention lapses.
+REFRESH_BUFFER_DAYS = 14
+# Meta requires tokens to be at least 24h old before they can be refreshed.
+MIN_TOKEN_AGE_HOURS = 24
+
+
+def load_env_lines() -> list[str]:
+    """Return the .env file as a list of raw lines (preserving comments +
+    blank lines), so we can rewrite individual keys without reformatting."""
+    if not IG_ENV_PATH.exists():
+        return []
+    return IG_ENV_PATH.read_text().splitlines()
+
+
+def env_dict_from_lines(lines: list[str]) -> dict[str, str]:
+    env = {}
+    for line in lines:
+        s = line.strip()
+        if not s or s.startswith("#") or "=" not in s:
+            continue
+        k, v = s.split("=", 1)
+        env[k.strip()] = v.strip()
+    return env
+
+
+def write_env_atomic(lines: list[str]):
+    """Rewrite the .env file from `lines`. Atomic via temp-file + os.replace
+    so a Ctrl-C or crash mid-write can't truncate the file."""
+    dir_ = IG_ENV_PATH.parent
+    dir_.mkdir(parents=True, exist_ok=True)
+    fd, tmp = tempfile.mkstemp(prefix=".env.tmp.", dir=str(dir_))
+    try:
+        with os.fdopen(fd, "w") as f:
+            f.write("\n".join(lines))
+            if lines and not lines[-1].endswith("\n"):
+                f.write("\n")
+        os.chmod(tmp, 0o600)
+        os.replace(tmp, IG_ENV_PATH)
+    finally:
+        if os.path.exists(tmp):
+            try:
+                os.unlink(tmp)
+            except OSError:
+                pass
+
+
+def expires_key_for(token_key: str) -> str:
+    """Derive the IG_TOKEN_EXPIRES env-var name from the IG_LONG_TOKEN one.
+
+    IG_LONG_TOKEN -> IG_TOKEN_EXPIRES
+    IG_LONG_TOKEN_MATTHEWHEARTFUL -> IG_TOKEN_EXPIRES_MATTHEWHEARTFUL
+    IG_LONG_TOKEN_OMIDOTME -> IG_TOKEN_EXPIRES_OMIDOTME
+    """
+    if not token_key.startswith("IG_LONG_TOKEN"):
+        return ""
+    return "IG_TOKEN_EXPIRES" + token_key[len("IG_LONG_TOKEN"):]
+
+
+def parse_expires(s: str | None) -> datetime | None:
+    if not s:
+        return None
+    s = s.strip()
+    if not s:
+        return None
+    # Accept both "2026-07-05T23:06:44Z" and "2026-07-05T23:06:44+00:00".
+    try:
+        if s.endswith("Z"):
+            s = s[:-1] + "+00:00"
+        return datetime.fromisoformat(s)
+    except ValueError:
+        return None
+
+
+def format_expires(dt: datetime) -> str:
+    """Match the existing .env convention: ISO-8601 UTC with trailing Z."""
+    return dt.astimezone(timezone.utc).strftime("%Y-%m-%dT%H:%M:%SZ")
+
+
+def refresh_token(long_token: str) -> dict:
+    qs = urllib.parse.urlencode({
+        "grant_type": "ig_refresh_token",
+        "access_token": long_token,
+    })
+    url = f"{GRAPH}/refresh_access_token?{qs}"
+    try:
+        with urllib.request.urlopen(url, timeout=30) as r:
+            return json.loads(r.read())
+    except urllib.error.HTTPError as e:
+        body = e.read().decode(errors="replace")
+        raise RefreshError(f"HTTP {e.code}: {body[:300]}") from e
+
+
+class RefreshError(Exception):
+    pass
+
+
+def update_line(lines: list[str], key: str, value: str) -> list[str]:
+    """Return a new list with the line `<key>=<old>` replaced by `<key>=<value>`.
+    If the key isn't present, appends `<key>=<value>` at the end."""
+    out = []
+    found = False
+    prefix = f"{key}="
+    for line in lines:
+        if line.strip().startswith(prefix) or line.startswith(prefix):
+            out.append(f"{key}={value}")
+            found = True
+        else:
+            out.append(line)
+    if not found:
+        out.append(f"{key}={value}")
+    return out
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--quiet", action="store_true")
+    parser.add_argument("--force", action="store_true",
+                        help="Refresh every token regardless of expiry buffer")
+    parser.add_argument("--dry-run", action="store_true",
+                        help="Print what would be refreshed but don't call the API")
+    parser.add_argument("--account", default=None,
+                        help="Only refresh this account (default: all accounts)")
+    args = parser.parse_args()
+
+    def log(msg: str):
+        if not args.quiet:
+            print(msg)
+
+    if not IG_ENV_PATH.exists():
+        print(f"[refresh-ig-tokens] env file missing: {IG_ENV_PATH}")
+        print("SUMMARY:REFRESHED=0 SKIPPED=0 FAILED=0 ACCOUNTS=0")
+        sys.exit(0)
+
+    try:
+        cfg = json.loads(SA_CONFIG.read_text())
+    except FileNotFoundError:
+        cfg = {}
+    accounts_cfg = ((cfg.get("instagram") or {}).get("accounts") or [])
+
+    if args.account:
+        accounts_cfg = [a for a in accounts_cfg
+                        if a.get("username", "").lower() == args.account.lower()]
+    if not accounts_cfg:
+        print("[refresh-ig-tokens] no instagram accounts in config")
+        print("SUMMARY:REFRESHED=0 SKIPPED=0 FAILED=0 ACCOUNTS=0")
+        sys.exit(0)
+
+    lines = load_env_lines()
+    env = env_dict_from_lines(lines)
+    now = datetime.now(timezone.utc)
+    buffer_secs = REFRESH_BUFFER_DAYS * 86400
+
+    refreshed = 0
+    skipped = 0
+    failed = 0
+
+    for account_cfg in accounts_cfg:
+        username = account_cfg.get("username", "")
+        token_key = account_cfg.get("ig_long_token_env", "IG_LONG_TOKEN")
+        exp_key = expires_key_for(token_key)
+        if not exp_key:
+            log(f"[refresh-ig-tokens] {username}: cannot derive expires key from {token_key}; skipping")
+            skipped += 1
+            continue
+
+        cur_token = env.get(token_key)
+        if not cur_token:
+            log(f"[refresh-ig-tokens] {username}: no value for {token_key}; skipping")
+            skipped += 1
+            continue
+
+        cur_exp_raw = env.get(exp_key)
+        cur_exp = parse_expires(cur_exp_raw)
+        if cur_exp is None and not args.force:
+            log(f"[refresh-ig-tokens] {username}: {exp_key} unparseable ({cur_exp_raw!r}); skipping (use --force to refresh anyway)")
+            skipped += 1
+            continue
+
+        if cur_exp is not None and not args.force:
+            remaining = (cur_exp - now).total_seconds()
+            if remaining > buffer_secs:
+                days_left = remaining / 86400
+                log(f"[refresh-ig-tokens] {username}: {days_left:.1f}d remaining (> {REFRESH_BUFFER_DAYS}d buffer); skipping")
+                skipped += 1
+                continue
+            if remaining < 0:
+                log(f"[refresh-ig-tokens] {username}: EXPIRED {(-remaining)/86400:.1f}d ago; attempting refresh anyway (Meta may reject)")
+
+        if args.dry_run:
+            log(f"[refresh-ig-tokens] {username}: DRY-RUN would refresh {token_key} (exp {cur_exp_raw})")
+            refreshed += 1
+            continue
+
+        log(f"[refresh-ig-tokens] {username}: refreshing {token_key} (current exp {cur_exp_raw})")
+        try:
+            resp = refresh_token(cur_token)
+        except RefreshError as e:
+            log(f"[refresh-ig-tokens] {username}: REFRESH FAILED: {e}")
+            failed += 1
+            continue
+        except Exception as e:
+            log(f"[refresh-ig-tokens] {username}: REFRESH FAILED (unexpected): {e}")
+            failed += 1
+            continue
+
+        new_token = resp.get("access_token")
+        expires_in = resp.get("expires_in")
+        if not new_token or not expires_in:
+            log(f"[refresh-ig-tokens] {username}: refresh response missing fields: {resp}")
+            failed += 1
+            continue
+
+        new_exp_dt = datetime.now(timezone.utc).fromtimestamp(time.time() + expires_in, tz=timezone.utc)
+        new_exp_str = format_expires(new_exp_dt)
+
+        lines = update_line(lines, token_key, new_token)
+        lines = update_line(lines, exp_key, new_exp_str)
+        env[token_key] = new_token
+        env[exp_key] = new_exp_str
+
+        log(f"[refresh-ig-tokens] {username}: OK, new expiry {new_exp_str} (~{expires_in/86400:.0f}d)")
+        refreshed += 1
+
+    if refreshed and not args.dry_run:
+        write_env_atomic(lines)
+        log(f"[refresh-ig-tokens] wrote {IG_ENV_PATH}")
+
+    print(
+        f"SUMMARY:REFRESHED={refreshed} SKIPPED={skipped} FAILED={failed} "
+        f"ACCOUNTS={len(accounts_cfg)}"
+    )
+
+
+if __name__ == "__main__":
+    main()

package/scripts/scan_instagram_comments.py

@@ -0,0 +1,481 @@
+#!/usr/bin/env python3
+"""Scan Instagram Graph API for new comments on our posts.
+
+For each enabled Instagram account in config.json (matt_diak, matthewheartful,
+omidotme), this:
+
+  1. Fetches /api/v1/posts?platform=instagram&our_account=<username> to build
+     a {shortcode: post_id} map of our DB-tracked IG posts.
+  2. Lists /me/media for the account (reuses the same Graph API call shape
+     update_instagram_stats.py uses).
+  3. For each media item present in our DB, calls /{media-id}/comments with
+     the replies sub-resource expanded.
+  4. Inserts each comment (and its nested replies) into the `replies` table
+     via reply_insert.insert_reply(). Server-side UNIQUE (platform,
+     their_comment_id) handles dedup; this script never SELECTs.
+
+Filters (mirrors scan_reddit_replies / scan_github_replies behavior):
+  - Skip comments whose author is in config.exclusions.authors
+  - Skip our own usernames (matt_diak / matthewheartful / omidotme) so we
+    don't try to reply to ourselves
+  - Skip backfill-old comments (older than BACKFILL_HOURS) with
+    status='skipped' / skip_reason='backfill_old'
+  - Skip too-short comments (< MIN_WORDS) with skip_reason='too_short'
+
+This is discovery-only. Posting replies back to Instagram lives in a separate
+engage script (Phase 2, not built yet); for now new rows surface in the
+dashboard replies feed as platform='instagram', status='pending'.
+
+Usage:
+    python3 scripts/scan_instagram_comments.py [--quiet] [--limit N]
+                                               [--account NAME]
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import os
+import sys
+import time
+import urllib.error
+import urllib.parse
+import urllib.request
+from datetime import datetime, timezone
+from pathlib import Path
+
+sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
+from http_api import api_get
+from reply_insert import insert_reply as _insert_reply
+
+
+IG_ENV_PATH = Path.home() / "instagram-graph-api" / ".env"
+GRAPH = "https://graph.instagram.com/v22.0"
+SA_CONFIG = Path(__file__).resolve().parent.parent / "config.json"
+
+# Discovery filters
+BACKFILL_HOURS = 48
+MIN_WORDS = 5
+# Per-Graph-API-call sleep so we stay polite under the 60/hr, 4800/day caps.
+# 3 accounts * ~10 media * (1 list + 1 comments call) = ~60 calls/cycle;
+# at 0.2s sleep that's ~12s per cycle, well inside 30-minute scheduling.
+GRAPH_SLEEP_SECS = 0.2
+
+
+# ── env / config ──────────────────────────────────────────────────────────────
+
+def load_ig_env() -> dict:
+    if not IG_ENV_PATH.exists():
+        return {}
+    env = {}
+    for line in IG_ENV_PATH.read_text().splitlines():
+        line = line.strip()
+        if not line or line.startswith("#") or "=" not in line:
+            continue
+        k, v = line.split("=", 1)
+        env[k.strip()] = v.strip()
+    return env
+
+
+def load_config() -> dict:
+    try:
+        return json.loads(SA_CONFIG.read_text())
+    except FileNotFoundError:
+        return {}
+
+
+def resolve_account_creds(account_name: str, ig_env: dict, accounts_cfg: list):
+    """Return (ig_user_id, long_token) or (None, None). Matches the lookup
+    pattern in scripts/update_instagram_stats.py."""
+    match = next(
+        (a for a in accounts_cfg if a.get("username", "").lower() == account_name.lower()),
+        None,
+    )
+    if match:
+        uid = ig_env.get(match.get("ig_user_id_env", "IG_USER_ID"))
+        tok = ig_env.get(match.get("ig_long_token_env", "IG_LONG_TOKEN"))
+        if uid and tok:
+            return uid, tok
+    uid = ig_env.get("IG_USER_ID")
+    tok = ig_env.get("IG_LONG_TOKEN")
+    return uid, tok
+
+
+# ── Graph API helpers ─────────────────────────────────────────────────────────
+
+def graph_get(path: str, token: str, **params):
+    params["access_token"] = token
+    url = f"{GRAPH}/{path}?{urllib.parse.urlencode(params)}"
+    with urllib.request.urlopen(url, timeout=20) as r:
+        return json.loads(r.read())
+
+
+def shortcode_from_url(url: str | None) -> str | None:
+    """Extract shortcode from an IG permalink.
+
+    https://www.instagram.com/reel/DYkkj8RDo9P/ -> DYkkj8RDo9P
+    """
+    import re
+    m = re.search(r"/(?:reel|p|tv)/([A-Za-z0-9_-]+)", url or "")
+    return m.group(1) if m else None
+
+
+def fetch_media_list(ig_user_id: str, token: str, max_pages: int = 5) -> list[dict]:
+    """Page through /me/media. Returns the raw items list with permalink + id."""
+    out = []
+    fields = "id,media_type,media_product_type,permalink,timestamp"
+    url = (
+        f"{GRAPH}/{ig_user_id}/media"
+        f"?fields={fields}&limit=100&access_token={token}"
+    )
+    pages = 0
+    while url and pages < max_pages:
+        with urllib.request.urlopen(url, timeout=20) as r:
+            data = json.loads(r.read())
+        out.extend(data.get("data", []) or [])
+        url = (data.get("paging") or {}).get("next")
+        pages += 1
+        if url:
+            time.sleep(GRAPH_SLEEP_SECS)
+    return out
+
+
+def fetch_comments(media_id: str, token: str) -> list[dict]:
+    """Return top-level comments for a media item, each with a nested
+    `replies.data[]` list (Graph API caps the sub-list at 25 by default; that
+    matches typical traffic on our posts)."""
+    fields = (
+        "id,username,text,timestamp,"
+        "replies{id,username,text,timestamp}"
+    )
+    try:
+        data = graph_get(f"{media_id}/comments", token, fields=fields, limit=50)
+    except urllib.error.HTTPError as e:
+        body = e.read().decode(errors="replace")[:200]
+        raise GraphApiError(f"HTTP {e.code} on /{media_id}/comments: {body}")
+    return data.get("data", []) or []
+
+
+class GraphApiError(Exception):
+    pass
+
+
+# ── posts lookup ──────────────────────────────────────────────────────────────
+
+def fetch_posts_map(account_username: str) -> dict[str, int]:
+    """Build {shortcode: post_id} for posts.platform='instagram' AND
+    posts.our_account=account_username. Uses the same /api/v1/posts endpoint
+    scan_reddit_replies.py uses for its post-id lookup."""
+    out: dict[str, int] = {}
+    resp = api_get(
+        "/api/v1/posts",
+        query={"platform": "instagram", "limit": 500},
+    )
+    posts = ((resp or {}).get("data") or {}).get("posts") or []
+    for p in posts:
+        if (p.get("our_account") or "").lower() != account_username.lower():
+            continue
+        code = shortcode_from_url(p.get("our_url"))
+        if code:
+            out[code] = int(p.get("id"))
+    return out
+
+
+# ── parse / classify ──────────────────────────────────────────────────────────
+
+def parse_ts(ts: str | None) -> float:
+    """Parse an IG ISO-8601 timestamp to a unix timestamp. Returns 0 on
+    failure (which counts as "old" for backfill purposes)."""
+    if not ts:
+        return 0.0
+    try:
+        # Instagram returns +0000 (no colon), strip and parse as UTC.
+        s = ts.replace("+0000", "+00:00")
+        return datetime.fromisoformat(s).timestamp()
+    except Exception:
+        return 0.0
+
+
+def word_count(text: str | None) -> int:
+    return len((text or "").split())
+
+
+def build_comment_url(shortcode: str, comment_id: str) -> str:
+    return f"https://www.instagram.com/p/{shortcode}/c/{comment_id}/"
+
+
+# ── main scan loop ────────────────────────────────────────────────────────────
+
+class IgCommentScanner:
+    def __init__(
+        self,
+        account_username: str,
+        ig_user_id: str,
+        token: str,
+        posts_map: dict[str, int],
+        excluded_authors: set[str],
+        quiet: bool = False,
+        media_limit: int | None = None,
+    ):
+        self.account = account_username
+        self.ig_user_id = ig_user_id
+        self.token = token
+        self.posts_map = posts_map
+        self.excluded = excluded_authors
+        self.quiet = quiet
+        self.media_limit = media_limit
+
+        self.discovered = 0
+        self.backfill_skipped = 0
+        self.too_short_skipped = 0
+        self.excluded_skipped = 0
+        self.already_tracked = 0
+        self.media_checked = 0
+        self.media_no_post = 0
+        self.comments_seen = 0
+
+    def log(self, msg: str):
+        if not self.quiet:
+            print(msg)
+
+    def _insert(
+        self,
+        post_id: int,
+        comment_id: str,
+        author: str,
+        content: str,
+        comment_url: str,
+        depth: int,
+        status: str,
+        skip_reason: str | None = None,
+    ):
+        result = _insert_reply(
+            None, post_id, "instagram", comment_id, author, content, comment_url,
+            parent_reply_id=None, depth=depth, status=status, skip_reason=skip_reason,
+        )
+        if result is None:
+            self.already_tracked += 1
+            return
+        if result == "pending":
+            self.discovered += 1
+        elif result == "skipped":
+            if skip_reason == "backfill_old":
+                self.backfill_skipped += 1
+            elif skip_reason and skip_reason.startswith("too_short"):
+                self.too_short_skipped += 1
+            elif skip_reason == "excluded_author":
+                self.excluded_skipped += 1
+
+    def _classify_and_insert(
+        self,
+        post_id: int,
+        shortcode: str,
+        comment: dict,
+        backfill_cutoff: float,
+        depth: int,
+    ):
+        comment_id = str(comment.get("id") or "")
+        if not comment_id:
+            return
+        self.comments_seen += 1
+        author = comment.get("username") or ""
+        content = comment.get("text") or ""
+        comment_url = build_comment_url(shortcode, comment_id)
+        created = parse_ts(comment.get("timestamp"))
+
+        if author.lower() in self.excluded:
+            self._insert(
+                post_id, comment_id, author, content, comment_url, depth,
+                status="skipped", skip_reason="excluded_author",
+            )
+            return
+
+        if created and created < backfill_cutoff:
+            self._insert(
+                post_id, comment_id, author, content, comment_url, depth,
+                status="skipped", skip_reason="backfill_old",
+            )
+            return
+
+        wc = word_count(content)
+        if wc < MIN_WORDS:
+            self._insert(
+                post_id, comment_id, author, content, comment_url, depth,
+                status="skipped", skip_reason=f"too_short ({wc} words)",
+            )
+            return
+
+        self._insert(
+            post_id, comment_id, author, content, comment_url, depth,
+            status="pending", skip_reason=None,
+        )
+
+    def scan(self):
+        self.log(f"[scan-ig-comments] account={self.account} posts_in_db={len(self.posts_map)}")
+        if not self.posts_map:
+            self.log(f"[scan-ig-comments]   no instagram posts in DB for account={self.account}; nothing to scan")
+            return
+
+        try:
+            media_items = fetch_media_list(self.ig_user_id, self.token)
+        except urllib.error.HTTPError as e:
+            body = e.read().decode(errors="replace")[:200]
+            self.log(f"[scan-ig-comments] /me/media failed for {self.account}: HTTP {e.code} {body}")
+            return
+        except Exception as e:
+            self.log(f"[scan-ig-comments] /me/media failed for {self.account}: {e}")
+            return
+
+        self.log(f"[scan-ig-comments]   /me/media returned {len(media_items)} items")
+        backfill_cutoff = time.time() - BACKFILL_HOURS * 3600
+
+        checked = 0
+        for item in media_items:
+            if self.media_limit and checked >= self.media_limit:
+                break
+            permalink = item.get("permalink")
+            shortcode = shortcode_from_url(permalink)
+            if not shortcode:
+                continue
+            post_id = self.posts_map.get(shortcode)
+            if not post_id:
+                self.media_no_post += 1
+                continue
+
+            media_id = item.get("id")
+            try:
+                comments = fetch_comments(media_id, self.token)
+            except GraphApiError as e:
+                self.log(f"[scan-ig-comments]   media={media_id} shortcode={shortcode} comments fetch failed: {e}")
+                continue
+
+            self.media_checked += 1
+            checked += 1
+            self.log(
+                f"[scan-ig-comments]   media={media_id} shortcode={shortcode} "
+                f"top_level_comments={len(comments)}"
+            )
+
+            for c in comments:
+                self._classify_and_insert(post_id, shortcode, c, backfill_cutoff, depth=1)
+                # Nested replies (replies to top-level comments). Author may
+                # be us (we already replied) or someone else (we got a reply
+                # to OUR reply). The excluded-author filter inside
+                # _classify_and_insert handles the first case.
+                replies = ((c.get("replies") or {}).get("data") or [])
+                for r in replies:
+                    self._classify_and_insert(post_id, shortcode, r, backfill_cutoff, depth=2)
+
+            time.sleep(GRAPH_SLEEP_SECS)
+
+    def summary(self) -> dict:
+        return {
+            "account": self.account,
+            "media_checked": self.media_checked,
+            "media_no_post_in_db": self.media_no_post,
+            "comments_seen": self.comments_seen,
+            "discovered": self.discovered,
+            "backfill_skipped": self.backfill_skipped,
+            "too_short_skipped": self.too_short_skipped,
+            "excluded_skipped": self.excluded_skipped,
+            "already_tracked": self.already_tracked,
+        }
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--quiet", action="store_true")
+    parser.add_argument("--limit", type=int, default=None,
+                        help="Cap media items inspected per account (debug)")
+    parser.add_argument("--account", default=None,
+                        help="Scan only this account (default: all enabled)")
+    args = parser.parse_args()
+
+    ig_env = load_ig_env()
+    cfg = load_config()
+    accounts_cfg = ((cfg.get("instagram") or {}).get("accounts") or [])
+    exclusions = cfg.get("exclusions") or {}
+    base_excluded = {a.lower() for a in (exclusions.get("authors") or [])}
+    # Always exclude our own usernames so we don't reply to ourselves.
+    own_usernames = {a.get("username", "").lower() for a in accounts_cfg if a.get("username")}
+
+    if args.account:
+        accounts_to_scan = [a for a in accounts_cfg
+                            if a.get("username", "").lower() == args.account.lower()]
+    else:
+        accounts_to_scan = [a for a in accounts_cfg if a.get("enabled", True)]
+
+    if not accounts_to_scan:
+        print("[scan-ig-comments] no instagram accounts to scan; exiting")
+        print("SUMMARY:DISCOVERED=0 SKIPPED=0 CHECKED=0 ALREADY=0 ACCOUNTS=0")
+        return
+
+    totals = {
+        "discovered": 0,
+        "backfill_skipped": 0,
+        "too_short_skipped": 0,
+        "excluded_skipped": 0,
+        "already_tracked": 0,
+        "media_checked": 0,
+        "comments_seen": 0,
+        "accounts": 0,
+    }
+
+    for account_cfg in accounts_to_scan:
+        username = account_cfg.get("username", "")
+        if not username:
+            continue
+        uid, tok = resolve_account_creds(username, ig_env, accounts_cfg)
+        if not uid or not tok:
+            print(f"[scan-ig-comments] missing creds for account={username}; skipping")
+            continue
+
+        excluded_for_account = set(base_excluded) | set(own_usernames)
+
+        try:
+            posts_map = fetch_posts_map(username)
+        except Exception as e:
+            print(f"[scan-ig-comments] posts lookup failed for {username}: {e}")
+            continue
+
+        scanner = IgCommentScanner(
+            username, uid, tok, posts_map, excluded_for_account,
+            quiet=args.quiet, media_limit=args.limit,
+        )
+        scanner.scan()
+        s = scanner.summary()
+        if not args.quiet:
+            print(
+                f"[scan-ig-comments] account={username} done: "
+                f"media_checked={s['media_checked']} comments_seen={s['comments_seen']} "
+                f"discovered={s['discovered']} "
+                f"backfill_skipped={s['backfill_skipped']} "
+                f"too_short_skipped={s['too_short_skipped']} "
+                f"excluded_skipped={s['excluded_skipped']} "
+                f"already_tracked={s['already_tracked']}"
+            )
+
+        totals["discovered"] += s["discovered"]
+        totals["backfill_skipped"] += s["backfill_skipped"]
+        totals["too_short_skipped"] += s["too_short_skipped"]
+        totals["excluded_skipped"] += s["excluded_skipped"]
+        totals["already_tracked"] += s["already_tracked"]
+        totals["media_checked"] += s["media_checked"]
+        totals["comments_seen"] += s["comments_seen"]
+        totals["accounts"] += 1
+
+    skipped_total = (
+        totals["backfill_skipped"]
+        + totals["too_short_skipped"]
+        + totals["excluded_skipped"]
+    )
+
+    print(
+        f"SUMMARY:DISCOVERED={totals['discovered']} SKIPPED={skipped_total} "
+        f"CHECKED={totals['media_checked']} ALREADY={totals['already_tracked']} "
+        f"ACCOUNTS={totals['accounts']}"
+    )
+
+
+if __name__ == "__main__":
+    main()

package/scripts/twitter_browser.py

@@ -609,8 +609,14 @@ def reply_to_tweet(tweet_url, text, apply_campaigns=True):
             except Exception:
                 pass
 
-            page.goto(tweet_url, wait_until="domcontentloaded")
-            page.wait_for_timeout(5000)
+            try:
+                page.goto(tweet_url, wait_until="load", timeout=60000)
+            except Exception:
+                try:
+                    page.goto(tweet_url, wait_until="domcontentloaded", timeout=60000)
+                except Exception:
+                    pass
+            page.wait_for_timeout(15000)
 
             # Check if page exists
             page_text = page.text_content("main") or ""
@@ -620,18 +626,20 @@ def reply_to_tweet(tweet_url, text, apply_campaigns=True):
             # Snapshot our reply links before posting (to detect the new one)
             links_before = _collect_our_reply_links(page)
 
-            # Find the reply textbox
+            # Find the reply textbox. On slower egress (E2B sandbox VMs) x.com
+            # can need 20-30s to attach the React reply composer; do not lower
+            # these timeouts.
             reply_box = None
             try:
                 reply_box = page.get_by_role("textbox", name="Post text")
-                reply_box.wait_for(timeout=10000)
+                reply_box.wait_for(timeout=30000)
             except Exception:
                 # Scroll down to find the reply box
                 page.evaluate("window.scrollBy(0, 500)")
-                page.wait_for_timeout(2000)
+                page.wait_for_timeout(3000)
                 try:
                     reply_box = page.get_by_role("textbox", name="Post text")
-                    reply_box.wait_for(timeout=5000)
+                    reply_box.wait_for(timeout=15000)
                 except Exception:
                     return {"ok": False, "error": "reply_box_not_found"}
 

package/skill/refresh-instagram-tokens.sh

@@ -0,0 +1,57 @@
+#!/usr/bin/env bash
+# refresh-instagram-tokens.sh — Refresh Instagram Graph API long-lived tokens
+# before they expire.
+#
+# IG long-lived tokens last ~60 days; this job runs daily and refreshes any
+# token within REFRESH_BUFFER_DAYS (default 14d) of expiry. The .env file at
+# ~/instagram-graph-api/.env is rewritten atomically on success.
+#
+# Lightweight (no lock needed — read+write to a file we own, no browser/MCP)
+# but we take instagram-poster anyway so a poster/stats/scan run that's mid-
+# flight can finish reading the existing token before we swap it.
+#
+# Logs: skill/logs/refresh-instagram-tokens-YYYY-MM-DD_HHMMSS.log
+
+set -uo pipefail
+
+REPO_DIR="$HOME/social-autoposter"
+LOG_DIR="$REPO_DIR/skill/logs"
+mkdir -p "$LOG_DIR"
+LOG_FILE="$LOG_DIR/refresh-instagram-tokens-$(date +%Y-%m-%d_%H%M%S).log"
+
+log() { echo "[$(date +%H:%M:%S)] $*" | tee -a "$LOG_FILE"; }
+log "=== refresh-instagram-tokens fire: $(date) ==="
+
+RUN_START=$(date +%s)
+
+# shellcheck source=lock.sh
+source "$REPO_DIR/skill/lock.sh"
+acquire_lock instagram-poster 30
+
+OUTPUT_FILE="/tmp/refresh-instagram-tokens-$$.out"
+if ! /opt/homebrew/bin/python3.11 "$REPO_DIR/scripts/refresh_instagram_tokens.py" 2>>"$LOG_FILE" | tee -a "$LOG_FILE" >"$OUTPUT_FILE"; then
+    log "refresh_instagram_tokens.py exited non-zero"
+    REFRESHED=0; SKIPPED=0; FAILED=0; ACCOUNTS=0
+else
+    SUMMARY=$(grep '^SUMMARY:' "$OUTPUT_FILE" | tail -1)
+    REFRESHED=$(echo "$SUMMARY" | sed -n 's/.*REFRESHED=\([0-9]*\).*/\1/p'); REFRESHED=${REFRESHED:-0}
+    SKIPPED=$(echo "$SUMMARY" | sed -n 's/.*SKIPPED=\([0-9]*\).*/\1/p'); SKIPPED=${SKIPPED:-0}
+    FAILED=$(echo "$SUMMARY" | sed -n 's/.*FAILED=\([0-9]*\).*/\1/p'); FAILED=${FAILED:-0}
+    ACCOUNTS=$(echo "$SUMMARY" | sed -n 's/.*ACCOUNTS=\([0-9]*\).*/\1/p'); ACCOUNTS=${ACCOUNTS:-0}
+fi
+rm -f "$OUTPUT_FILE"
+
+RUN_ELAPSED=$(( $(date +%s) - RUN_START ))
+
+log "logging run: refreshed=$REFRESHED skipped=$SKIPPED failed=$FAILED accounts=$ACCOUNTS elapsed=${RUN_ELAPSED}s"
+
+/opt/homebrew/bin/python3.11 "$REPO_DIR/scripts/log_run.py" \
+    --script "refresh_instagram_tokens" \
+    --posted "$REFRESHED" \
+    --skipped "$SKIPPED" \
+    --failed "$FAILED" \
+    --cost 0 \
+    --elapsed "$RUN_ELAPSED" >>"$LOG_FILE" 2>&1 || log "log_run.py failed"
+
+log "=== refresh-instagram-tokens done ==="
+exit 0

package/skill/run-instagram-daily.sh

@@ -38,8 +38,28 @@ fi
 
 log() { echo "[$(date +%H:%M:%S)] $*" | tee -a "$LOG_FILE"; }
 
+# Run accounting for dashboard Job History (Post Threads · Instagram).
+# Each exit site updates POSTED_CT / SKIPPED_CT / FAILED_CT; the EXIT trap
+# always emits one log_run.py line so the run shows up under
+# thread_instagram, matching how thread_twitter / thread_reddit log.
+RUN_START_EPOCH=$(date +%s)
+POSTED_CT=0
+SKIPPED_CT=0
+FAILED_CT=0
+
 cleanup() {
+  local rc=$?
   rm -f "$PICK_FILE"
+  if [ "$POSTED_CT" -eq 0 ] && [ "$SKIPPED_CT" -eq 0 ] && [ "$FAILED_CT" -eq 0 ]; then
+    if [ "$rc" -eq 0 ]; then SKIPPED_CT=1; else FAILED_CT=1; fi
+  fi
+  local elapsed=$(( $(date +%s) - RUN_START_EPOCH ))
+  local cost
+  cost=$(/usr/bin/python3 "$REPO_DIR/scripts/get_run_cost.py" --since "$RUN_START_EPOCH" --scripts "run-instagram-daily" 2>/dev/null || echo "0.0000")
+  /usr/bin/python3 "$REPO_DIR/scripts/log_run.py" \
+      --script "thread_instagram" \
+      --posted "$POSTED_CT" --skipped "$SKIPPED_CT" --failed "$FAILED_CT" \
+      --cost "$cost" --elapsed "$elapsed" >/dev/null 2>&1 || true
 }
 trap cleanup EXIT INT TERM HUP
 
@@ -65,6 +85,7 @@ else
 fi
 if [ -z "$TARGET_ACCOUNT" ]; then
   log "pick_ig_account.py produced no account — exiting non-zero"
+  FAILED_CT=1
   exit 1
 fi
 log "picker chose account: $TARGET_ACCOUNT"
@@ -76,9 +97,11 @@ if ! /opt/homebrew/bin/python3.11 "$REPO_DIR/scripts/ig_post_type_picker.py" \
   rc=$?
   if [ "$rc" -eq 2 ]; then
     log "queue exhausted for account=$TARGET_ACCOUNT (no drafts of either type) — exiting cleanly"
+    SKIPPED_CT=1
     exit 0
   fi
   log "picker failed rc=$rc — exiting non-zero"
+  FAILED_CT=1
   exit 1
 fi
 
@@ -93,6 +116,7 @@ log "picker reason: ${REASON}"
 
 if [ ! -f "$VIDEO_PATH" ]; then
   log "ERROR: picker pointed at $VIDEO_PATH but file missing on disk"
+  FAILED_CT=1
   exit 1
 fi
 
@@ -107,9 +131,11 @@ log "step 3: post_to_ig.py --file $(basename "$VIDEO_PATH") --post-type $POST_TY
 if ! /opt/homebrew/bin/python3.11 "$REPO_DIR/mixer/post_to_ig.py" \
         --file "$VIDEO_PATH" --post-type "$POST_TYPE" --account "$TARGET_ACCOUNT" $DRY_FLAG >>"$LOG_FILE" 2>&1; then
   log "post_to_ig.py failed — exiting non-zero"
+  FAILED_CT=1
   exit 1
 fi
 
+POSTED_CT=1
 log "=== finished post-${POST_NUMBER} (${POST_TYPE}) on ${TARGET_ACCOUNT} successfully ==="
 
 # Step 4: mirror the new media_posts row into the cross-platform `posts` table

package/skill/run-twitter-cycle.sh

@@ -1047,7 +1047,7 @@ export CLAUDE_SESSION_ID
 
 PREP_SCHEMA='{"type":"object","properties":{"candidates":{"type":"array","items":{"type":"object","properties":{"candidate_id":{"type":"integer"},"candidate_url":{"type":"string"},"thread_author":{"type":"string"},"thread_text":{"type":"string"},"matched_project":{"type":"string"},"reply_text":{"type":"string"},"engagement_style":{"type":"string"},"language":{"type":"string"},"has_landing_pages":{"type":"boolean"},"link_keyword":{"type":"string"},"link_slug":{"type":"string"}},"required":["candidate_id","candidate_url","matched_project","reply_text","engagement_style","language","has_landing_pages"]}},"rejected":{"type":"array","items":{"type":"object","properties":{"candidate_id":{"type":"integer"},"reason":{"type":"string"},"proposed_excludes":{"type":"array","items":{"type":"string"}}},"required":["candidate_id","reason"]}}},"required":["candidates","rejected"]}'
 
-PREP_OUTPUT=$("$REPO_DIR/scripts/run_claude.sh" "run-twitter-cycle-prep" --strict-mcp-config --mcp-config "$TW_MCP_CONFIG" -p --output-format json --json-schema "$PREP_SCHEMA" "${TW_ENGINE_PREFIX}You are the Social Autoposter prep step.
+PREP_PROMPT="${TW_ENGINE_PREFIX}You are the Social Autoposter prep step.
 
 Your ONLY job in THIS session:
   1. Read each thread you decide to reply to (browser tools from the BROWSER BACKEND block above, READ-ONLY).
@@ -1130,7 +1130,13 @@ CRITICAL:
 - DO NOT call log_post.py or campaign_bump.py.
 - Browser tools (from the BROWSER BACKEND block) are READ-ONLY in this step.
 - NEVER use em dashes. Use commas, periods, or regular dashes (-).
-- Reply in the SAME LANGUAGE as the parent tweet." 2>&1)
+- Reply in the SAME LANGUAGE as the parent tweet."
+
+# Pipe the prep prompt via stdin instead of passing as a shell argument.
+# On Linux ARG_MAX is 2MB; the assembled prompt (config.json + top_report +
+# styles + schema + candidates) busts that on the VM, dying with E2BIG
+# "Argument list too long". stdin has no such cap.
+PREP_OUTPUT=$(printf '%s' "$PREP_PROMPT" | "$REPO_DIR/scripts/run_claude.sh" "run-twitter-cycle-prep" --strict-mcp-config --mcp-config "$TW_MCP_CONFIG" -p --output-format json --json-schema "$PREP_SCHEMA" 2>&1)
 
 echo "$PREP_OUTPUT" >> "$LOG_FILE"
 

package/skill/scan-instagram-replies.sh

@@ -0,0 +1,61 @@
+#!/usr/bin/env bash
+# scan-instagram-replies.sh — Discover new inbound comments on our Instagram
+# posts via the Graph API and insert them into the `replies` table.
+#
+# Mirrors the pattern used by stats-instagram.sh: API-only (no browser),
+# instagram-poster lock (so scan, stats, and post can't race for the same
+# token-bucket), then a SUMMARY-line parsed by log_run.py for the dashboard
+# Jobs panel.
+#
+# Logs: skill/logs/scan-instagram-replies-YYYY-MM-DD_HHMMSS.log
+
+set -uo pipefail
+
+REPO_DIR="$HOME/social-autoposter"
+LOG_DIR="$REPO_DIR/skill/logs"
+mkdir -p "$LOG_DIR"
+LOG_FILE="$LOG_DIR/scan-instagram-replies-$(date +%Y-%m-%d_%H%M%S).log"
+
+log() { echo "[$(date +%H:%M:%S)] $*" | tee -a "$LOG_FILE"; }
+log "=== scan-instagram-replies fire: $(date) ==="
+
+RUN_START=$(date +%s)
+
+# instagram-poster lock — stats, scan, daily-post, and render all share this
+# lane so we don't race on the same /me/media token bucket.
+# shellcheck source=lock.sh
+source "$REPO_DIR/skill/lock.sh"
+acquire_lock instagram-poster 30
+
+OUTPUT_FILE="/tmp/scan-instagram-replies-$$.out"
+if ! /opt/homebrew/bin/python3.11 "$REPO_DIR/scripts/scan_instagram_comments.py" 2>>"$LOG_FILE" | tee -a "$LOG_FILE" >"$OUTPUT_FILE"; then
+    log "scan_instagram_comments.py exited non-zero — logging run as failed"
+    DISCOVERED=0; SKIPPED=0; CHECKED=0; ALREADY=0; ACCOUNTS=0
+else
+    SUMMARY=$(grep '^SUMMARY:' "$OUTPUT_FILE" | tail -1)
+    DISCOVERED=$(echo "$SUMMARY" | sed -n 's/.*DISCOVERED=\([0-9]*\).*/\1/p'); DISCOVERED=${DISCOVERED:-0}
+    SKIPPED=$(echo "$SUMMARY" | sed -n 's/.*SKIPPED=\([0-9]*\).*/\1/p'); SKIPPED=${SKIPPED:-0}
+    CHECKED=$(echo "$SUMMARY" | sed -n 's/.*CHECKED=\([0-9]*\).*/\1/p'); CHECKED=${CHECKED:-0}
+    ALREADY=$(echo "$SUMMARY" | sed -n 's/.*ALREADY=\([0-9]*\).*/\1/p'); ALREADY=${ALREADY:-0}
+    ACCOUNTS=$(echo "$SUMMARY" | sed -n 's/.*ACCOUNTS=\([0-9]*\).*/\1/p'); ACCOUNTS=${ACCOUNTS:-0}
+fi
+rm -f "$OUTPUT_FILE"
+
+RUN_ELAPSED=$(( $(date +%s) - RUN_START ))
+
+log "logging run: discovered=$DISCOVERED skipped=$SKIPPED checked=$CHECKED already=$ALREADY accounts=$ACCOUNTS elapsed=${RUN_ELAPSED}s"
+
+# discovered -> posted (new pending rows are the productive output of a scan,
+# same convention scan_reddit_replies / scan_github_replies use).
+# skipped -> skipped. checked -> scanned (media items inspected).
+/opt/homebrew/bin/python3.11 "$REPO_DIR/scripts/log_run.py" \
+    --script "scan_instagram_comments" \
+    --posted "$DISCOVERED" \
+    --skipped "$SKIPPED" \
+    --failed 0 \
+    --scanned "$CHECKED" \
+    --cost 0 \
+    --elapsed "$RUN_ELAPSED" >>"$LOG_FILE" 2>&1 || log "log_run.py failed"
+
+log "=== scan-instagram-replies done ==="
+exit 0