@m13v/s4l 1.6.197-rc.7

package/skill/lib/twitter-backend.sh

@@ -0,0 +1,314 @@
+#!/bin/bash
+# twitter-backend.sh - Twitter pipeline browser bootstrap (harness-only since
+# 2026-05-19; the legacy twitter-agent Playwright MCP path was fully ripped out).
+#
+# Source this AFTER lock.sh, BEFORE any acquire_lock / browser pre-flight /
+# claude -p subprocess calls. Sets these for the caller:
+#
+#   MCP_CONFIG_FILE        - claude -p --mcp-config path (twitter-harness MCP)
+#   BROWSER_INSTRUCTIONS   - prompt block describing the harness backend +
+#                            its bh_run tool surface (inject at the TOP of any
+#                            prompt that mentions browser_* tools)
+#
+# And exports (so Python subprocesses like twitter_browser.py inherit them):
+#
+#   TWITTER_CDP_URL        - http://127.0.0.1:9555 (forces direct CDP attach,
+#                            skipping ps-based agent-profile discovery)
+#
+# Provides these functions (names preserved for back-compat with existing
+# callers in engage-twitter.sh, run-twitter-cycle.sh, run-twitter-threads.sh,
+# dm-outreach-twitter.sh, scan-twitter-followups.sh):
+#
+#   ensure_twitter_browser_for_backend
+#     Call AFTER acquire_lock "twitter-browser". Probes harness Chrome on
+#     port 9555 and launches it idempotently if down, then cleans leftover
+#     tabs from prior runs.
+#
+#   defer_if_foreign_for_backend [log_file]
+#     No-op. Harness CDP supports multiple concurrent clients on the same
+#     Chrome (no SingletonLock fight), so foreign MCP wrappers never block
+#     us. Kept as a function only so callers don't have to change.
+
+MCP_CONFIG_FILE="$HOME/.claude/browser-agent-configs/twitter-harness-mcp.json"
+
+# Per-host env override (written by bin/cli.js when installing on an AppMaker
+# VM, where the canonical browser is Chromium on port 9222 behind the SOAX
+# residential proxy at 127.0.0.1:3003, NOT the harness Chrome on 9555). On a
+# Mac dev box this file does not exist, so the default below kicks in.
+if [ -f "$HOME/.social-autoposter-env" ]; then
+    # shellcheck disable=SC1091
+    . "$HOME/.social-autoposter-env"
+fi
+
+# Tell twitter_browser.py (and any other Python helper that honors this env
+# var) to skip ps-based discovery and connect directly to the configured CDP
+# endpoint. Default 9555 (Mac harness Chrome). AppMaker VMs pre-set this to
+# http://127.0.0.1:9222 via ~/.social-autoposter-env above.
+export TWITTER_CDP_URL="${TWITTER_CDP_URL:-http://127.0.0.1:9555}"
+
+# Default harness URL — used by ensure_twitter_browser_for_backend +
+# cleanup_harness_tabs to decide whether we own this Chrome (and should
+# launch/clean it) or whether it is externally managed (AppMaker, BYO).
+_BH_DEFAULT_URL="http://127.0.0.1:9555"
+# DEPRECATED 2026-06-26: this block is NO LONGER injected into any model prompt.
+# run-twitter-cycle.sh now sets TW_ENGINE_PREFIX="" — Phase 1 (query) and Phase 2b
+# (prep) are tool-free; the model drafts from inlined candidate context only, and all
+# browser work is the shell's deterministic CDP scan + Phase 2b-post's
+# twitter_browser.py. Kept only so ensure_twitter_browser_for_backend's existing
+# assignment doesn't break; safe to delete once confirmed unreferenced. Do NOT
+# reintroduce the "logged in as m13v_" hardcode or a model-facing bh_run contract.
+BROWSER_INSTRUCTIONS=$(cat <<'BROWSER_HARNESS_EOF'
+BROWSER BACKEND: twitter-harness (browser-harness MCP, CDP-driven REAL Google Chrome on
+port 9555, profile ~/.claude/browser-profiles/browser-harness). The Chrome is already
+logged in as m13v_; cookies persist on disk.
+
+You have ONE tool: mcp__twitter-harness__bh_run(script). It runs arbitrary Python with
+these helpers pre-imported:
+  new_tab(url), goto_url(url), wait_for_load(), page_info(),
+  capture_screenshot(),                     # returns path to PNG; Read it to see the page
+  click_at_xy(x, y),                        # coordinate click (viewport pixels)
+  js(expression),                           # page.evaluate-style; returns the result
+  type_text(text),                          # types into currently-focused element
+  press_key(key),                           # e.g. "Enter", "Tab", "Escape"
+  scroll(direction, amount), cdp(method, **params)
+
+TAB HYGIENE (IMPORTANT): A placeholder tab ALWAYS already exists when you start
+(pre-flight leaves exactly one tab open). REUSE IT: use goto_url() for your VERY FIRST
+navigation as well as every subsequent one, so the existing tab is navigated in place.
+Call new_tab() ONLY as a fallback when no usable tab exists (goto_url errors because
+there is no active page) OR when you genuinely need a second tab open in parallel.
+Opening a fresh tab on first navigation orphans the placeholder and leaks a tab every
+cycle, which exhausts per-process Chrome resources.
+
+TRANSLATION TABLE - wherever this prompt mentions a Playwright-style tool, do the
+following with bh_run instead:
+
+  browser_navigate(url)           ->  Reuse the existing tab (default, incl. first nav):
+                                       bh_run('goto_url("URL"); wait_for_load()')
+                                       Fallback only if no tab exists / parallel tab needed:
+                                       bh_run('new_tab("URL"); wait_for_load()')
+  browser_snapshot                ->  bh_run('print(js("""..."""))') to read DOM as structured data,
+                                       OR bh_run('print(capture_screenshot())') + Read the PNG
+  browser_run_code(js)            ->  bh_run('print(js("""<the JS expression>"""))')
+  browser_click(ref=...)          ->  Find the element via selector, compute center coords from
+                                       getBoundingClientRect, then bh_run('click_at_xy(X, Y)')
+  browser_type(ref=..., text=...) ->  Click the textbox first (click_at_xy), then bh_run('type_text("TEXT")')
+  browser_take_screenshot         ->  bh_run('print(capture_screenshot())') then Read the path
+  browser_press_key("Enter")      ->  bh_run('press_key("Enter")')
+
+EXAMPLE - click the reply submit button:
+  bh_run('''
+  pt = js("""
+    const el = document.querySelector('[data-testid="tweetButtonInline"]');
+    if (!el) return null;
+    const r = el.getBoundingClientRect();
+    return {x: r.x + r.width/2, y: r.y + r.height/2};
+  """)
+  print(pt)
+  ''')
+  # Then in a follow-up call (substituting the x/y from above):
+  bh_run('click_at_xy(123, 456)')
+
+VERIFY AFTER EVERY MUTATION by capturing a screenshot and reading the PNG, coordinate
+clicks can miss; visual verification is the only reliable confirmation that the action took.
+BROWSER_HARNESS_EOF
+)
+
+cleanup_harness_tabs() {
+    # Close every CDP "page" tab except one. Delegated to a standalone Python
+    # script because bash 3.2 (what launchd uses) cannot parse a nested heredoc
+    # inside a function body inside a sourced file. Inline form here broke every
+    # launchd-fired twitter script on 2026-05-14 until this refactor.
+    #
+    # Health-check gate: 2026-05-16 the original `--max-time 2` was too strict.
+    # When harness Chrome is busy (long scans, lock backups, CPU-pinned),
+    # the /json/version probe times out, cleanup is silently skipped, and the
+    # next scan's new_tab() leaks an orphan tab. Symptom: occasional
+    # "closed 14/14 extra page tabs" cycles after several skips piled up.
+    # Now: 10s timeout + ONE retry; log skips so they are not silent.
+    local _probe="curl -sf --max-time 10 -o /dev/null http://127.0.0.1:9555/json/version"
+    if ! $_probe 2>/dev/null; then
+        sleep 1
+        if ! $_probe 2>/dev/null; then
+            echo "[$(date +%H:%M:%S)] cleanup_harness_tabs: SKIPPED (harness CDP /json/version unreachable after 10s+retry)" >&2
+            return 0
+        fi
+    fi
+    python3 "$HOME/social-autoposter/scripts/cleanup_harness_tabs.py" 2>/dev/null || true
+}
+
+_resolve_chrome_bin() {
+    # Auto-detect Chrome/Chromium so the same script launches the harness on
+    # macOS dev boxes AND Linux VMs. Override with BH_CHROME_BIN.
+    if [ -n "${BH_CHROME_BIN:-}" ] && [ -x "$BH_CHROME_BIN" ]; then
+        echo "$BH_CHROME_BIN"; return 0
+    fi
+    for _p in \
+        "/Applications/Google Chrome.app/Contents/MacOS/Google Chrome" \
+        "/Applications/Chromium.app/Contents/MacOS/Chromium" \
+        "/usr/bin/google-chrome" "/usr/bin/google-chrome-stable" \
+        "/usr/bin/chromium" "/usr/bin/chromium-browser" "/snap/bin/chromium"
+    do
+        if [ -x "$_p" ]; then echo "$_p"; return 0; fi
+    done
+    for _n in google-chrome google-chrome-stable chromium chromium-browser; do
+        _which=$(command -v "$_n" 2>/dev/null) && [ -n "$_which" ] && { echo "$_which"; return 0; }
+    done
+    echo ""; return 1
+}
+
+ensure_twitter_browser_for_backend() {
+    # AppMaker / BYO Chrome: TWITTER_CDP_URL points at something other than our
+    # default harness URL. Don't touch that browser; just probe it and bail.
+    # The AppMaker bootstrap (and any future BYO setup) is responsible for
+    # keeping the externally-managed Chrome alive.
+    if [ "${TWITTER_CDP_URL:-$_BH_DEFAULT_URL}" != "$_BH_DEFAULT_URL" ]; then
+        local _ext_url="${TWITTER_CDP_URL}"
+        if curl -sf --max-time 2 -o /dev/null "${_ext_url}/json/version" 2>/dev/null; then
+            echo "[$(date +%H:%M:%S)] Using externally-managed Chrome at ${_ext_url} (skipping harness launch + tab cleanup)" >&2
+            # Restore the Twitter login if the sandbox was substituted. AppMaker
+            # Hobby-tier sandboxes have a 1h TTL; on substitution /root is reseeded
+            # from /etc/skel-root and the harness profile (cookies) is wiped. This
+            # re-injects the stored session from social_accounts via the HTTP API.
+            # No-op when already logged in. Never blocks the cycle on failure.
+            python3 "$HOME/social-autoposter/scripts/restore_twitter_session.py" 2>&1 | sed 's/^/[restore] /' >&2 || true
+            return 0
+        fi
+        echo "[$(date +%H:%M:%S)] ERROR: TWITTER_CDP_URL=${_ext_url} not reachable. External Chrome must be managed by host (AppMaker /opt/startup.sh, etc)." >&2
+        return 1
+    fi
+    # Probe + launch harness Chrome on port 9555 if needed.
+    if ! curl -sf --max-time 2 -o /dev/null http://127.0.0.1:9555/json/version 2>/dev/null; then
+        echo "[$(date +%H:%M:%S)] Harness Chrome down on port 9555, launching..." >&2
+        local _chrome_bin
+        _chrome_bin=$(_resolve_chrome_bin)
+        if [ -z "$_chrome_bin" ]; then
+            echo "[$(date +%H:%M:%S)] ERROR: no Chrome/Chromium binary found. Set BH_CHROME_BIN." >&2
+            return 1
+        fi
+        # On Linux + no display, run headless. On root, add --no-sandbox.
+        # Window-position/size only meaningful on macOS multi-monitor; skip
+        # elsewhere so we don't hide the window off-screen on single-display
+        # Linux VMs.
+        local _extra=()
+        case "$(uname -s)" in
+            Linux)
+                _extra+=(--no-sandbox --disable-dev-shm-usage)
+                if [ -z "${DISPLAY:-}" ] && [ -z "${WAYLAND_DISPLAY:-}" ]; then
+                    _extra+=(--headless=new --disable-gpu)
+                fi
+                ;;
+            Darwin)
+                _extra+=(--window-position="${BH_WINDOW_POS:-3042,-1032}")
+                _extra+=(--window-size="${BH_WINDOW_SIZE:-1024,1013}")
+                ;;
+        esac
+        # --password-store=basic + --use-mock-keychain: encrypt the cookie store
+        # with Chrome's fixed obfuscation key instead of the macOS Keychain
+        # ("Chrome Safe Storage"). Without this, a keychain lock/re-lock leaves
+        # Chrome unable to decrypt its Cookies SQLite on the next launch, so it
+        # discards the session and the harness comes up logged out. With it, the
+        # x.com cookies persist + decrypt across restarts natively, no
+        # re-injection needed. Matches the flags the Playwright browser agents
+        # already use. (Root-cause persistence fix, 2026-06-02; the cookie
+        # mirror + restore_twitter_session.py remain as the safety net.)
+        # Self-heal (2026-06-03): if a Chrome already holds THIS profile dir but
+        # is not answering CDP on our port, a fresh launch hands off to it via
+        # Chrome's SingletonLock and exits without ever binding our port — the
+        # old "failed to start within 12s" loop (8h Twitter outage overnight
+        # 2026-06-02/03, root cause: a server.py regression that dropped
+        # BH_PROFILE_NAME and collapsed the linkedin/twitter harness profiles
+        # onto this one, stranding an orphan on 9556). Reap the stale owner of
+        # our EXACT profile dir (trailing space in the pattern so browser-harness
+        # never matches browser-harness-linkedin) before relaunching.
+        local _prof_dir="$HOME/.claude/browser-profiles/browser-harness"
+        local _stale_pids
+        _stale_pids=$(pgrep -f -- "--user-data-dir=$_prof_dir " 2>/dev/null || true)
+        if [ -n "$_stale_pids" ] && ! curl -sf --max-time 2 -o /dev/null http://127.0.0.1:9555/json/version 2>/dev/null; then
+            echo "[$(date +%H:%M:%S)] CDP down but Chrome still holds $_prof_dir (pids: $(echo $_stale_pids | tr '\n' ' ')); reaping stale profile owner before relaunch" >&2
+            kill $_stale_pids 2>/dev/null || true
+            sleep 2
+            _stale_pids=$(pgrep -f -- "--user-data-dir=$_prof_dir " 2>/dev/null || true)
+            [ -n "$_stale_pids" ] && { kill -9 $_stale_pids 2>/dev/null || true; sleep 1; }
+            rm -f "$_prof_dir/SingletonLock" "$_prof_dir/SingletonSocket" "$_prof_dir/SingletonCookie" 2>/dev/null || true
+        fi
+        "$_chrome_bin" \
+            --remote-debugging-port=9555 \
+            --user-data-dir="$HOME/.claude/browser-profiles/browser-harness" \
+            --no-first-run --no-default-browser-check \
+            --password-store=basic --use-mock-keychain \
+            --disable-features=ChromeWhatsNewUI,CalculateNativeWinOcclusion \
+            --disable-backgrounding-occluded-windows \
+            "${_extra[@]}" \
+            "${BH_LAUNCH_URL:-https://x.com}" >/dev/null 2>&1 &
+        disown
+        for _i in 1 2 3 4 5 6 7 8 9 10 11 12; do
+            curl -sf --max-time 2 -o /dev/null http://127.0.0.1:9555/json/version 2>/dev/null && break
+            sleep 1
+        done
+        if ! curl -sf --max-time 2 -o /dev/null http://127.0.0.1:9555/json/version 2>/dev/null; then
+            echo "[$(date +%H:%M:%S)] ERROR: harness Chrome failed to start within 12s" >&2
+            return 1
+        fi
+        echo "[$(date +%H:%M:%S)] Harness Chrome up on port 9555" >&2
+    fi
+    # Re-inject the stored X session if the harness Chrome is logged out — e.g. a
+    # keychain re-lock wiped Chrome's encrypted Cookies SQLite on this launch
+    # (Gap B, 2026-06-02). restore_twitter_session.py reads the keychain-
+    # independent local cookie mirror (written by connect_x) and injects via CDP.
+    # No-op when already logged in; never blocks the cycle on failure. Runs on
+    # both the freshly-launched and already-up paths so a mid-life logout heals.
+    TWITTER_CDP_URL="http://127.0.0.1:9555" \
+        python3 "$HOME/social-autoposter/scripts/restore_twitter_session.py" 2>&1 \
+        | sed 's/^/[restore] /' >&2 || true
+    # Always close leftover tabs from prior runs. Safe under acquire_lock
+    # "twitter-browser" serialization (every caller of this function holds
+    # that lock), so we will not race with another active twitter run.
+    cleanup_harness_tabs
+}
+
+defer_if_foreign_for_backend() {
+    # Harness Chrome accepts multiple concurrent CDP clients on the same
+    # browser-harness profile, so a foreign MCP wrapper (Fazm Dev / IDE)
+    # cannot cause the SingletonLock contention that historically blocked
+    # the twitter-agent profile. Always return 1 (do not defer).
+    return 1
+}
+
+# --- browser-harness `-c` capability self-heal (added 2026-06-02) -----------
+# A stale ~/Developer/browser-harness checkout that PREDATES the `-c` interface
+# makes `browser-harness -c "<script>"` print its usage string instead of
+# running the script. The Phase 1 scan loop in run-twitter-cycle.sh then yields
+# zero tweets with no obvious cause. cli.js documents the same failure for the
+# bh_run MCP path. When this bit the testing machine, the debugging agent saw
+# the `-c` flag, WRONGLY assumed it was unsupported, and proposed rewriting the
+# call to a nonexistent "stdin form" (browser-harness has no stdin mode — `-c`
+# is the only interface; see run.py). This runs at source-time, before any
+# `-c` call, so all twitter harness scripts (cycle/threads/engage/dm/followups)
+# get auto-repair. Static probe is one grep when fresh (zero steady-state cost);
+# the git+uv refresh only fires when the checkout is actually stale.
+_sa_harness_log() {
+    # Use the caller's log() FUNCTION when present; `declare -F` matches only a
+    # shell function, never the macOS /usr/bin/log binary (command -v would).
+    if declare -F log >/dev/null 2>&1; then log "$*"; else echo "[$(date +%H:%M:%S)] $*" >&2; fi
+}
+_sa_resolve_uv() {
+    local c
+    c="$(command -v uv 2>/dev/null)" && { echo "$c"; return 0; }
+    for c in "$HOME/.local/bin/uv" /opt/homebrew/bin/uv /usr/local/bin/uv; do
+        [ -x "$c" ] && { echo "$c"; return 0; }
+    done
+    return 1
+}
+ensure_harness_c_support() {
+    # Retired 2026-06-02. Upstream browser-harness removed `-c` in favor of
+    # stdin-heredoc (commits after merge-base 0e679e2); our server.py wrapper
+    # now passes scripts via stdin (input=script) so the CLI shape doesn't
+    # need any pre-flight probing. The old gate grepped run.py for `"-c"`
+    # which always fails against current upstream, and its "self-heal" was a
+    # `git reset --hard FETCH_HEAD` on ~/Developer/browser-harness that
+    # would clobber local commits AND not actually re-add `-c`. Keep the
+    # name + no-op return so older sourced contexts that call it don't break.
+    return 0
+}
+ensure_harness_c_support || true

package/skill/link-edit-github.sh

@@ -0,0 +1,136 @@
+#!/usr/bin/env bash
+# link-edit-github.sh — Edit existing GitHub issue comments to append a project link.
+# Uses the gh CLI (no browser needed). GitHub has no upvote system, so eligibility
+# is based on our comment being posted 6h+ ago (engagement = a reply in the issue thread).
+# Called by launchd (com.m13v.social-link-edit-github) every 6 hours.
+
+set -euo pipefail
+
+# Cycle ID for cross-cycle cost accounting (see run-github.sh for the same
+# pattern). Stamps claude_sessions.cycle_id via env inheritance.
+BATCH_ID="${BATCH_ID:-legh-$(date +%Y%m%d-%H%M%S)}"
+export BATCH_ID
+export SA_CYCLE_ID="$BATCH_ID"
+
+# Platform lock: wait up to 45min for any previous link-edit-github run, then skip
+source "$(dirname "$0")/lock.sh"
+acquire_lock "link-edit-github" 2700
+
+# Load secrets
+# shellcheck source=/dev/null
+[ -f "$HOME/social-autoposter/.env" ] && source "$HOME/social-autoposter/.env"
+
+REPO_DIR="$HOME/social-autoposter"
+SKILL_FILE="$REPO_DIR/SKILL.md"
+LOG_DIR="$REPO_DIR/skill/logs"
+# HTTP-only lane (2026-06-01): all reads/writes go through the s4l.ai API via
+# scripts/link_edit_helper.py. No DATABASE_URL, no psql, no fallback.
+LE_HELPER="$REPO_DIR/scripts/link_edit_helper.py"
+
+mkdir -p "$LOG_DIR"
+LOG_FILE="$LOG_DIR/link-edit-github-$(date +%Y-%m-%d_%H%M%S).log"
+
+log() { echo "[$(date +%H:%M:%S)] $*" | tee -a "$LOG_FILE"; }
+
+RUN_START=$(date +%s)
+log "=== GitHub Link Edit Run: $(date) ==="
+
+# A/B gate: per-post deterministic coin flip for the page-gen lane. Mirrors
+# scripts/twitter_gen_links.py's TWITTER_PAGE_GEN_RATE behavior and the
+# Reddit link-edit pipeline's LINK_EDIT_REDDIT_PAGE_GEN_RATE. 0.30 means
+# ~30% of eligible posts get a brand-new SEO landing page built inline via
+# Claude + git push; the other ~70% fall through to the project's homepage
+# with link_source='plain_url_ab_skip'. Per-post hash via Postgres
+# hashtext() so the same post stays in the same lane across cron retries.
+# Tunable via env var so cadence sweeps don't need code changes. 0.0
+# disables page-gen entirely (link insertion still happens with plain URL);
+# 1.0 restores 100% page-gen.
+# DEFAULT 0.0: GitHub no longer generates custom SEO pages — every eligible
+# post goes through the wrap-an-existing-link route (homepage + short link).
+LINK_EDIT_GITHUB_PAGE_GEN_RATE="${LINK_EDIT_GITHUB_PAGE_GEN_RATE:-0.0}"
+PAGE_GEN_RATE_PCT=$(python3 -c "v=float('$LINK_EDIT_GITHUB_PAGE_GEN_RATE'); v=max(0.0,min(1.0,v)); print(int(round(v*100)))")
+log "A/B gate: LINK_EDIT_GITHUB_PAGE_GEN_RATE=$LINK_EDIT_GITHUB_PAGE_GEN_RATE (page_gen_lane='page_gen' on ~${PAGE_GEN_RATE_PCT}% of eligible posts; rest go to plain_url_ab_skip)"
+
+EDITABLE=$(python3 "$LE_HELPER" eligible --platform github --page-gen-rate-pct "$PAGE_GEN_RATE_PCT" --order posted_at 2>/dev/null || echo "")
+
+if [ "$EDITABLE" = "null" ] || [ -z "$EDITABLE" ]; then
+    log "No GitHub posts eligible for link edit"
+    python3 "$REPO_DIR/scripts/log_run.py" --script "link_edit_github" --posted 0 --skipped 0 --failed 0 --cost 0 --elapsed $(( $(date +%s) - RUN_START ))
+    exit 0
+fi
+
+EDITABLE_COUNT=$(echo "$EDITABLE" | python3 -c "import json,sys; print(len(json.load(sys.stdin)))" 2>/dev/null || echo "?")
+log "GitHub: $EDITABLE_COUNT posts eligible for link edit"
+
+PROMPT_FILE=$(mktemp)
+cat > "$PROMPT_FILE" <<PROMPT_EOF
+You are the Social Autoposter GitHub link-edit bot.
+
+Read $SKILL_FILE for the full workflow. Execute the GitHub link-edit phase only. GitHub edits are done via the gh CLI (no browser). GitHub has no upvote system; engagement = someone replied to our comment in the issue thread.
+
+CRITICAL: This is a single-shot run. NEVER call ScheduleWakeup, CronCreate, CronDelete, CronList, EnterPlanMode, EnterWorktree, or any deferred-execution / scheduling tool. You MUST complete or skip every post in this one run; do not defer work to "a future run". If you hit a hard block, mark the post SKIPPED via step 9 and move on to the next post.
+
+GitHub posts eligible for editing:
+$EDITABLE
+
+Process ALL of them. For each post:
+1. Read ~/social-autoposter/config.json to get the projects list.
+2. Pick the project whose topics are the CLOSEST match to thread_title + our_content. Check the project_name column first; if set, use that project directly. Otherwise match by topics. Be generous: if the thread touches agents, automation, desktop, memory, or anything related to the project descriptions, it's a match. If truly nothing fits, mark it skipped (see step 9) and move on. Frame it as recommending a cool tool you've come across, NOT as something you built.
+3. PAGE-GEN LANE GATE — read the post's \`page_gen_lane\` field (set deterministically by the pipeline; do NOT override).
+   - If \`page_gen_lane == "ab_skip"\`: SKIP the full SEO page generation entirely. Set LINK_URL = the matched project's homepage from config.json (the \`website\` field) and LINK_SOURCE="plain_url_ab_skip". Continue to step 4. The /r/<code> short-link wrap in step 5 still mints attribution on the project's own domain, so we get click data for this lane to compare against seo_page lane CTR.
+   - If \`page_gen_lane == "page_gen"\` AND the matched project has a landing_pages config: continue to step 3a below.
+   - If \`page_gen_lane == "page_gen"\` BUT the matched project has NO landing_pages config: skip page-gen, set LINK_URL = project homepage (website if available, otherwise github), LINK_SOURCE="plain_url_no_lp", continue to step 4.
+
+3a. If the matched project has a landing_pages config (with repo, base_url):
+   a. Think about what SEO-optimized guide page would fit this specific thread naturally. Consider the thread's audience, their pain points, industry jargon, and what they'd actually find useful. The page should NOT feel like a landing page; it should feel like a genuine 1000-2000 word guide or resource.
+   b. cd into the project repo (landing_pages.repo)
+   c. Look at existing pages under src/app/t/ to understand the site's style, layout components (Navbar, Footer), and theme
+   d. Create a NEW standalone page as src/app/t/{seo-friendly-slug}/page.tsx; this is a real Next.js page with its own Metadata export, not a JSON entry. Include:
+      - Proper <Metadata> with title, description, openGraph, twitter tags
+      - Reuse the site's Navbar and Footer components (import or inline them)
+      - Use the CTAButton component from @/components/cta-button for ALL call-to-action buttons (it tracks clicks in PostHog automatically). Import: import { CTAButton } from "@/components/cta-button";
+      - A full article-style page: hero headline, table of contents, 5-7 content sections, comparison tables with real numbers, bullet lists with specific data points, and a CTA section at the bottom
+      - The content must be 1000-2000 words. Pull real context from the project's config (pricing, features, proof_points, competitive_positioning) and from web research to make it concrete and authoritative
+      - Naturally mention the product as ONE solution among the options discussed; don't make the whole page a sales pitch
+   e. git add the new page && git commit -m "Add guide: SHORT_DESCRIPTION" && git push
+   f. Wait ~35s for Vercel deploy, then curl -sI {base_url}/t/{slug} to verify HTTP 200
+   g. On success, set LINK_URL = the deployed page URL and LINK_SOURCE="seo_page". On deploy failure, fall back GRACEFULLY: set LINK_URL = the project's homepage from config.json (the \`website\` field), set LINK_SOURCE="plain_url_fallback:deploy_failed". Do NOT skip the post; continue to step 4.
+4. Write 1 sentence + project link (GitHub peer tone). Voice depends on the matched project's \`voice_relationship\` field in config.json (read it before drafting):
+   - voice_relationship == "first_party": Claim ownership. Examples: "fwiw we built an implementation of this, URL" or "I shipped a tool that does this, URL". NEVER write "I found this", "there's a tool", "came across this implementation".
+   - voice_relationship == "third_party": You are an outside observer pointing at the project's mechanism. Example: "fwiw PROJECT_NAME has an implementation of this, URL". Do NOT use "I built" / "we shipped" / "we made". Do NOT use "I found this" / "came across this" either; stay matter-of-fact.
+5. URL-WRAP THE LINK TEXT for click attribution. Run:
+     python3 ~/social-autoposter/scripts/dm_short_links.py wrap-post-text \\
+       --text "YOUR_LINK_SENTENCE_WITH_URL" \\
+       --platform github_issues \\
+       --project PROJECT_NAME
+   Parse the JSON output. Use \`text\` (URL replaced with /r/<code>) as the FINAL LINK_TEXT for steps 6 and 7. Keep \`minted_session\` for step 8. If wrap returns ok=false, log the error and skip this post (do NOT post a raw URL).
+6. Extract OWNER/REPO from thread_url. Extract COMMENT_ID from our_url; if not directly available, use gh api to find our comment on that issue.
+7. Edit the existing comment (append the wrapped LINK_TEXT to the existing content) using gh:
+   gh api repos/OWNER/REPO/issues/comments/COMMENT_ID -X PATCH -f body="EXISTING_CONTENT
+
+   WRAPPED_LINK_TEXT"
+8. After each successful edit, update the DB (via the HTTP API helper; pass link_source so we can A/B compare seo_page vs plain_url_ab_skip vs plain_url_fallback:* vs plain_url_no_lp click-through rates, same as Twitter does in scripts/twitter_gen_links.py and the Reddit link-edit pipeline does) and backfill short-link attribution:
+   python3 ~/social-autoposter/scripts/link_edit_helper.py mark-edited --post-id POST_ID --content "LINK_TEXT" --source "LINK_SOURCE"
+   python3 ~/social-autoposter/scripts/dm_short_links.py backfill-post --minted-session MINTED_SESSION --post-id POST_ID
+9. COMMITMENT GUARDRAILS (never violate these):
+   - NEVER suggest, offer, or agree to calls, meetings, demos, or video chats.
+   - NEVER promise to share links, files, or resources you don't have right now. Only share links from config.json projects (plus any new landing page you just deployed).
+   - NEVER offer to DM or send anything outside the comment.
+   - NEVER make time-bound promises.
+10. If a post is SKIPPED (no project match, comment not found, issue locked, 404, bad URL), ALWAYS mark it so it won't be retried:
+    python3 ~/social-autoposter/scripts/link_edit_helper.py mark-skipped --post-id POST_ID --reason "REASON"
+PROMPT_EOF
+
+gtimeout 1800 "$REPO_DIR/scripts/run_claude.sh" "link-edit-github" --strict-mcp-config --mcp-config "$HOME/.claude/browser-agent-configs/no-agents-mcp.json" --disallowed-tools "ScheduleWakeup,CronCreate,CronDelete,CronList,EnterPlanMode,EnterWorktree" --output-format stream-json --verbose -p "$(cat "$PROMPT_FILE")" 2>&1 | tee -a "$LOG_FILE" || log "WARNING: GitHub link-edit claude exited with code $?"
+rm -f "$PROMPT_FILE"
+
+EDITED=$(python3 "$LE_HELPER" edited-count --platform github 2>/dev/null || echo "0")
+log "GitHub link-edit complete. Total github posts edited (all-time): $EDITED"
+
+RUN_ELAPSED=$(( $(date +%s) - RUN_START ))
+_COST=$(python3 "$REPO_DIR/scripts/get_run_cost.py" --since "$RUN_START" --scripts "link-edit-github" 2>/dev/null || echo "0.0000")
+python3 "$REPO_DIR/scripts/log_run.py" --script "link_edit_github" --posted 0 --skipped 0 --failed 0 --cost "$_COST" --elapsed "$RUN_ELAPSED"
+
+find "$LOG_DIR" -name "link-edit-github-*.log" -mtime +7 -delete 2>/dev/null || true
+
+log "=== GitHub link-edit complete: $(date) ==="

package/skill/link-edit-moltbook.sh

@@ -0,0 +1,117 @@
+#!/usr/bin/env bash
+# link-edit-moltbook.sh — Edit high-performing Moltbook comments to append a project link.
+# Moltbook uses the PATCH API (no browser needed).
+# Called by launchd (com.m13v.social-link-edit-moltbook) every 6 hours.
+
+set -euo pipefail
+
+# Cycle ID for cross-cycle cost accounting (see run-moltbook.sh for the same
+# pattern). Stamps claude_sessions.cycle_id via env inheritance.
+BATCH_ID="${BATCH_ID:-lemb-$(date +%Y%m%d-%H%M%S)}"
+export BATCH_ID
+export SA_CYCLE_ID="$BATCH_ID"
+
+# Platform lock: wait up to 45min for any previous link-edit-moltbook run, then skip
+source "$(dirname "$0")/lock.sh"
+acquire_lock "link-edit-moltbook" 2700
+
+# Load secrets
+# shellcheck source=/dev/null
+[ -f "$HOME/social-autoposter/.env" ] && source "$HOME/social-autoposter/.env"
+
+REPO_DIR="$HOME/social-autoposter"
+SKILL_FILE="$REPO_DIR/SKILL.md"
+LOG_DIR="$REPO_DIR/skill/logs"
+# HTTP-only lane (2026-06-01): all reads/writes go through the s4l.ai API via
+# scripts/link_edit_helper.py. No DATABASE_URL, no psql, no fallback.
+LE_HELPER="$REPO_DIR/scripts/link_edit_helper.py"
+
+mkdir -p "$LOG_DIR"
+LOG_FILE="$LOG_DIR/link-edit-moltbook-$(date +%Y-%m-%d_%H%M%S).log"
+
+log() { echo "[$(date +%H:%M:%S)] $*" | tee -a "$LOG_FILE"; }
+
+RUN_START=$(date +%s)
+log "=== Moltbook Link Edit Run: $(date) ==="
+
+EDITABLE=$(python3 "$LE_HELPER" eligible --platform moltbook --min-upvotes-exclusive 2 --order upvotes 2>/dev/null || echo "")
+
+if [ "$EDITABLE" = "null" ] || [ -z "$EDITABLE" ]; then
+    log "No Moltbook posts eligible for link edit"
+    python3 "$REPO_DIR/scripts/log_run.py" --script "link_edit_moltbook" --posted 0 --skipped 0 --failed 0 --cost 0 --elapsed $(( $(date +%s) - RUN_START ))
+    exit 0
+fi
+
+EDITABLE_COUNT=$(echo "$EDITABLE" | python3 -c "import json,sys; print(len(json.load(sys.stdin)))" 2>/dev/null || echo "?")
+log "Moltbook: $EDITABLE_COUNT posts eligible for link edit"
+
+PROMPT_FILE=$(mktemp)
+cat > "$PROMPT_FILE" <<PROMPT_EOF
+You are the Social Autoposter Moltbook link-edit bot.
+
+Read $SKILL_FILE for the full workflow. Execute the Moltbook link-edit phase only. Moltbook uses the PATCH API; no browser is needed.
+
+CRITICAL: This is a single-shot run. NEVER call ScheduleWakeup, CronCreate, CronDelete, CronList, EnterPlanMode, EnterWorktree, or any deferred-execution / scheduling tool. You MUST complete or skip every post in this one run; do not defer work to "a future run". If you hit a hard block, mark the post SKIPPED via step 9 and move on to the next post.
+
+Moltbook posts eligible for editing:
+$EDITABLE
+
+Process ALL of them. For each post:
+1. Read ~/social-autoposter/config.json to get the projects list.
+2. Pick the project whose topics are the CLOSEST match to thread_title + our_content. Check the project_name column first; if set, use that project directly. Otherwise match by topics. Be generous: if the thread touches agents, automation, desktop, memory, or anything related to the project descriptions, it's a match. If truly nothing fits, mark it skipped (see step 8) and move on. Frame it as recommending a cool tool you've come across, NOT as something you built.
+3. If the matched project has a landing_pages config (with repo, base_url):
+   a. Think about what SEO-optimized guide page would fit this specific thread naturally. Consider the thread's audience, their pain points, industry jargon, and what they'd actually find useful. The page should NOT feel like a landing page; it should feel like a genuine 1000-2000 word guide or resource.
+   b. cd into the project repo (landing_pages.repo)
+   c. Look at existing pages under src/app/t/ to understand the site's style, layout components (Navbar, Footer), and theme
+   d. Create a NEW standalone page as src/app/t/{seo-friendly-slug}/page.tsx; this is a real Next.js page with its own Metadata export, not a JSON entry. Include:
+      - Proper <Metadata> with title, description, openGraph, twitter tags
+      - Reuse the site's Navbar and Footer components (import or inline them)
+      - Use the CTAButton component from @/components/cta-button for ALL call-to-action buttons (it tracks clicks in PostHog automatically). Import: import { CTAButton } from "@/components/cta-button";
+      - A full article-style page: hero headline, table of contents, 5-7 content sections, comparison tables with real numbers, bullet lists with specific data points, and a CTA section at the bottom
+      - The content must be 1000-2000 words. Pull real context from the project's config (pricing, features, proof_points, competitive_positioning) and from web research to make it concrete and authoritative
+      - Naturally mention the product as ONE solution among the options discussed; don't make the whole page a sales pitch
+   e. git add the new page && git commit -m "Add guide: SHORT_DESCRIPTION" && git push
+   f. Wait ~35s for Vercel deploy, then curl -sI {base_url}/t/{slug} to verify HTTP 200
+   g. Use THAT page URL in the link edit. If deploy fails, fall back to the project's website URL.
+   If no landing_pages config: use website if available, otherwise github.
+4. Write 1 casual sentence + project link (Moltbook agent voice). Voice depends on the matched project's \`voice_relationship\` field in config.json (read it before drafting):
+   - voice_relationship == "first_party": Claim ownership. Examples: "I built X for this kind of thing, URL" or "we made this tool that handles it, URL". NEVER write "there's this cool tool", "I found this", "came across this".
+   - voice_relationship == "third_party": You are an outside observer pointing at the project's mechanism. Example: "PROJECT_NAME handles this kind of thing, URL". Do NOT use "I built" / "we made". Do NOT use "I found this" / "came across this" either; stay matter-of-fact.
+5. URL-WRAP THE LINK TEXT for click attribution. Run:
+     python3 ~/social-autoposter/scripts/dm_short_links.py wrap-post-text \\
+       --text "YOUR_LINK_SENTENCE_WITH_URL" \\
+       --platform moltbook \\
+       --project PROJECT_NAME
+   Parse the JSON output. Use \`text\` (URL replaced with /r/<code>) as the FINAL LINK_TEXT for steps 6 and 7. Keep \`minted_session\` for step 8. If wrap returns ok=false, log the error and skip this post (do NOT post a raw URL).
+6. Append the wrapped LINK_TEXT to our_content with a blank line separator.
+7. Extract the comment UUID from our_url (the part after #comment-), then PATCH the comment:
+   source ~/social-autoposter/.env
+   curl -s -X PATCH -H "Authorization: Bearer \$MOLTBOOK_API_KEY" \\
+     -H "Content-Type: application/json" \\
+     -d '{"content": "FULL_CONTENT"}' \\
+     "https://www.moltbook.com/api/v1/comments/COMMENT_UUID"
+8. After each successful edit, update the DB (via the HTTP API helper) and backfill short-link attribution:
+   python3 ~/social-autoposter/scripts/link_edit_helper.py mark-edited --post-id POST_ID --content "LINK_TEXT"
+   python3 ~/social-autoposter/scripts/dm_short_links.py backfill-post --minted-session MINTED_SESSION --post-id POST_ID
+9. COMMITMENT GUARDRAILS (never violate these):
+   - NEVER suggest, offer, or agree to calls, meetings, demos, or video chats.
+   - NEVER promise to share links, files, or resources you don't have right now. Only share links from config.json projects (plus any new landing page you just deployed).
+   - NEVER offer to DM or send anything outside the comment.
+   - NEVER make time-bound promises.
+10. If a post is SKIPPED (no project match, comment not found, removed, bad URL), ALWAYS mark it so it won't be retried:
+    python3 ~/social-autoposter/scripts/link_edit_helper.py mark-skipped --post-id POST_ID --reason "REASON"
+PROMPT_EOF
+
+gtimeout 1800 "$REPO_DIR/scripts/run_claude.sh" "link-edit-moltbook" --strict-mcp-config --mcp-config "$HOME/.claude/browser-agent-configs/no-agents-mcp.json" --disallowed-tools "ScheduleWakeup,CronCreate,CronDelete,CronList,EnterPlanMode,EnterWorktree" -p "$(cat "$PROMPT_FILE")" 2>&1 | tee -a "$LOG_FILE" || log "WARNING: Moltbook link-edit claude exited with code $?"
+rm -f "$PROMPT_FILE"
+
+EDITED=$(python3 "$LE_HELPER" edited-count --platform moltbook 2>/dev/null || echo "0")
+log "Moltbook link-edit complete. Total moltbook posts edited (all-time): $EDITED"
+
+RUN_ELAPSED=$(( $(date +%s) - RUN_START ))
+_COST=$(python3 "$REPO_DIR/scripts/get_run_cost.py" --since "$RUN_START" --scripts "link-edit-moltbook" 2>/dev/null || echo "0.0000")
+python3 "$REPO_DIR/scripts/log_run.py" --script "link_edit_moltbook" --posted 0 --skipped 0 --failed 0 --cost "$_COST" --elapsed "$RUN_ELAPSED"
+
+find "$LOG_DIR" -name "link-edit-moltbook-*.log" -mtime +7 -delete 2>/dev/null || true
+
+log "=== Moltbook link-edit complete: $(date) ==="