@m13v/s4l 1.7.1-rc.2 → 1.7.1-rc.21

package/scripts/twitter_post_plan.py

@@ -424,10 +424,11 @@ def post_one(c: dict, picker_assignment: dict | None = None) -> tuple[str, str]:
     outcome: 'posted' | 'skipped' | 'failed'
     reason:  short failure key when outcome != 'posted', else ''.
 
-    picker_assignment: optional {assigned_style, assigned_mode} dict
-        sourced from the plan envelope. When present, drives the
-        validate_or_register call below so USE-mode drift coerces back
-        and INVENT-mode new_style blocks land in
+    picker_assignment: optional {assigned_style, assigned_mode} dict for
+        THIS candidate (per-candidate override when the plan carries one,
+        else the plan-level fallback, see call site in main()). When
+        present, drives the validate_or_register call below so USE-mode
+        drift coerces back and INVENT-mode new_style blocks land in
         engagement_styles_registry. None means legacy behaviour
         (uncoerced; whatever the model said is what gets logged).
     """
@@ -1031,13 +1032,14 @@ def main() -> int:
         os.environ["CLAUDE_SESSION_ID"] = plan_session_id
 
     # Pull the picker assignment from the plan envelope (written by
-    # run-twitter-cycle.sh after s4l_pick_style). Shared across every
-    # candidate in the batch because the picker fires once per cycle.
-    # Falls back to None on legacy plans (pre-2026-05-22 envelopes that
-    # don't carry these keys); post_one then runs the legacy uncoerced
-    # path. Empty assigned_style + assigned_mode='invent' means the
-    # picker rolled INVENT this cycle; validate_or_register treats that
-    # as "register if the model produced a well-formed new_style block".
+    # run-twitter-cycle.sh after s4l_pick_style). This is the PLAN-LEVEL
+    # fallback, used verbatim for legacy plans (pre-2026-05-22 envelopes, or
+    # any candidate missing a per-candidate assignment) and as the last
+    # resort in the per-candidate override below. Falls back to None on
+    # legacy plans that don't carry these keys; post_one then runs the
+    # legacy uncoerced path. Empty assigned_style + assigned_mode='invent'
+    # means the picker rolled INVENT this cycle; validate_or_register treats
+    # that as "register if the model produced a well-formed new_style block".
     picker_assignment = {
         "assigned_style": plan.get("assigned_style") or None,
         "assigned_mode":  plan.get("assigned_mode")  or None,
@@ -1169,7 +1171,25 @@ def main() -> int:
                 except Exception:
                     pass
             try:
-                outcome, reason = post_one(c, picker_assignment=picker_assignment)
+                # Two-draft cards (2026-07-07): each candidate may carry its OWN
+                # (assigned_style, assigned_mode) reflecting whichever draft is
+                # actually posting, either the recommended one (stamped at
+                # plan-write time) or the other one (stamped by the MCP
+                # post_drafts edits path when a human switched drafts on the
+                # review card). Without this, every card would coerce back to
+                # ONE cycle-wide style even when it posted under the other
+                # style, silently corrupting the engagement_style label the
+                # picker's performance stats are learned from. Falls back to
+                # the plan-level assignment for any candidate/plan that
+                # predates this (assigned_mode key absent).
+                if "assigned_mode" in c:
+                    _cand_assignment = {
+                        "assigned_style": c.get("assigned_style"),
+                        "assigned_mode": c.get("assigned_mode"),
+                    }
+                else:
+                    _cand_assignment = picker_assignment
+                outcome, reason = post_one(c, picker_assignment=_cand_assignment)
             except Exception as e:
                 print(f"[post] candidate {c.get('candidate_id')} crashed: {e}",
                       flush=True)
@@ -1247,7 +1267,13 @@ def main() -> int:
                 "twitter post pipeline issues: "
                 f"posted={posted} failed={failed} attempted={len(candidates)} "
                 f"reasons={','.join(f'{k}:{v}' for k, v in machine.items())}",
-                level=("error" if (failed > 0 or posted == 0) else "warning"),
+                # error only when a REAL failure happened (failed > 0). A batch
+                # that's all benign operational skips (tweet_not_found because
+                # the target vanished before we replied, rate_limited, etc.)
+                # with posted==0 is not a problem, it's a batch of 1 that had
+                # nothing postable. Conflating "nothing posted" with "broken"
+                # used to file a fresh Sentry error for every such skip.
+                level=("error" if failed > 0 else "warning"),
                 tags={
                     "component": "twitter_post",
                     "posted": str(posted),

package/skill/dm-outreach-twitter.sh

@@ -146,6 +146,25 @@ For each DM row, BEFORE you compose or send, do this in order:
    - 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".
+   - MANDATORY BLOCK CHECK (do this on every profile visit, not just when a
+     send later fails): read the page text (snapshot or DOM query) and check
+     for the literal string "has blocked you" (case-insensitive). This is the
+     ONLY sufficient evidence for concluding we're blocked -- never infer it
+     from a missing Message button, a screenshot, or a send failure alone
+     (a missing Message button just as often means "DMs closed", which is
+     chat_disabled below, not a block).
+     If that exact text IS present:
+       a. Skip this DM: python3 $REPO_DIR/scripts/dm_conversation.py mark-skipped \\
+            --dm-id DM_ID --reason "blocked_by_user"
+       b. Hard-block the author so future items from them are never drafted
+          (shared universal blocklist, same mechanism engage-twitter uses):
+            python3 $REPO_DIR/scripts/reply_db.py blocklist add x THEIR_AUTHOR \\
+              --reason "X reported this author has blocked our account (dm-outreach-twitter, profile check)" \\
+              --classification blocked_by_author \\
+              --added-by block_probe
+       c. Move on to the next DM row; do not attempt to send.
+     If that text is NOT present, proceed normally -- do not skip or block
+     based on any other signal from this step.
 
 3. Persist the profile fields:
    \`\`\`bash

package/skill/engage-twitter.sh

@@ -356,6 +356,33 @@ MANDATORY reply flow for every item:
           a) Re-snapshot the page to refresh element state.
           b) Find the reply textbox: role="textbox" with name like "Post your reply"
              or "Post text". Click it.
+             If the textbox is NOT found, do NOT assume a technical glitch or
+             conclude "blocked" from a screenshot alone -- X removes the
+             composer but still renders the tweet normally when the author has
+             blocked us, so a missing textbox is genuinely ambiguous by itself.
+             Determine which deterministically:
+               i.   Click the article's reply icon (data-testid="reply" on the
+                    target article, NOT the submit button).
+               ii.  Wait ~2s, then read the page text (snapshot or DOM query).
+               iii. If the text contains "has blocked you" (case-insensitive):
+                    this is a confirmed, permanent block. Do NOT retry. Skip
+                    this reply: python3 $REPO_DIR/scripts/reply_db.py skipped ID "author_blocked_us"
+                    Then hard-block the author so future items from them are
+                    never drafted (same mechanism as the escape hatch above):
+                      python3 $REPO_DIR/scripts/reply_db.py blocklist add x HANDLE \\
+                        --reason "X reported this author has blocked our account (live reply attempt, engage-twitter)" \\
+                        --classification blocked_by_author \\
+                        --added-by block_probe \\
+                        --source-reply-id REPLY_ID
+                    Move on to the next assigned reply.
+               iv.  If that text is NOT present: this was a transient render
+                    miss, not a block. The click in (i) may have opened the
+                    composer -- re-snapshot; if the textbox is now present,
+                    continue with it below. If still absent, retry this whole
+                    step once more before treating it as a failed technical
+                    attempt.
+             NEVER conclude "blocked" from a screenshot alone -- only the
+             literal "has blocked you" text match above is sufficient evidence.
           c) Type YOUR_REPLY_TEXT (post-Step-3b short-link-wrapped form, post-Step-3a
              suffix) into that textbox. Do NOT auto-submit; we click the Reply
              button explicitly in step e.

package/skill/run-draft-and-publish.sh

@@ -123,20 +123,21 @@ if [ -n "${S4L_FORCE_PROJECT:-}" ]; then
     echo "[run-draft-and-publish] personal_brand mode: forcing project '$S4L_FORCE_PROJECT' (link-free)" >&2
 fi
 
-# Draft-only flag (2026-07-06). Per-cycle DRAFT_ONLY decision, overriding the
-# plist's baked DRAFT_ONLY=1 baseline:
-#   - draft-only ON (default), or persona cycle -> DRAFT_ONLY=1: draft stops
-#     before posting and the plan merges into review cards, exactly as before.
-#   - draft-only OFF + promotion cycle -> DRAFT_ONLY=0: the cycle POSTS its
-#     top-1 pick autonomously; the rolling virality bar activates automatically
-#     as the no-human quality gate.
-# S4L_CYCLE_LANE comes from the s4l_mode env eval above; when absent (e.g. a
-# personal_brand-only setup with no persona provisioned yet) we default to the
-# safe card path.
+# Draft-only flag (2026-07-06, made lane-uniform 2026-07-08). Per-cycle
+# DRAFT_ONLY decision, overriding the plist's baked DRAFT_ONLY=1 baseline:
+#   - draft-only ON (default): DRAFT_ONLY=1 for every lane; draft stops before
+#     posting and the plan merges into review cards, exactly as before.
+#   - draft-only OFF: DRAFT_ONLY=0 for every lane (promotion AND
+#     personal_brand); the cycle POSTS its top-1 pick autonomously. The
+#     rolling virality bar activates automatically as the no-human quality
+#     gate for promotion cycles.
+# Previously this only applied to the promotion lane (personal_brand always
+# forced DRAFT_ONLY=1). Operator instruction 2026-07-08: draft-only must be a
+# single global switch that covers every lane, not just promotion.
 DRAFT_ONLY_FLAG="$("$PY" "$REPO_DIR/scripts/s4l_mode.py" draft-only 2>/dev/null || echo 1)"
-if [ "${S4L_CYCLE_LANE:-}" = "promotion" ] && [ "$DRAFT_ONLY_FLAG" = "0" ]; then
+if [ "$DRAFT_ONLY_FLAG" = "0" ]; then
     export DRAFT_ONLY=0
-    echo "[run-draft-and-publish] draft-only OFF: promotion cycle will POST autonomously (DRAFT_ONLY=0, virality bar active)" >&2
+    echo "[run-draft-and-publish] draft-only OFF: cycle (lane=${S4L_CYCLE_LANE:-unknown}) will POST autonomously (DRAFT_ONLY=0)" >&2
 else
     export DRAFT_ONLY=1
 fi

package/skill/run-linkedin.sh

@@ -1094,9 +1094,17 @@ echo "Phase A: chose project=$PA_PROJECT topic='$PA_SEARCH_TOPIC' activity=$PA_A
 
 # Look up the chosen project's full config (only this one).
 PROJECT_FULL=$(python3 -c "
-import json, os
+import json, os, sys
+sys.path.insert(0, os.path.expanduser('~/social-autoposter/scripts'))
+import learned_preferences as lp
 c = json.load(open(os.path.expanduser('~/social-autoposter/config.json')))
 p = next((p for p in c.get('projects',[]) if p['name']=='$PA_PROJECT'), {})
+# learned_preferences is a SINGLE install-wide block since 2026-07-08 (see
+# scripts/learned_preferences.py), not one per project. Stamped in under the
+# same key so the existing prose below ('learned_preferences.draft_style_notes
+# entry ... overrides the engagement style') keeps working unchanged.
+p = dict(p)
+p['learned_preferences'] = lp.get_global_block(c)
 print(json.dumps(p, indent=2))
 ")
 

package/skill/run-twitter-cycle.sh

@@ -1748,9 +1748,17 @@ Project match: $cproject${DRAFT_LINE}${AUTHOR_HISTORY_LINE}
 done <<< "$CANDIDATES"
 
 ALL_PROJECTS_JSON=$(python3 -c "
-import json, os
+import json, os, sys
+sys.path.insert(0, os.path.expanduser('~/social-autoposter/scripts'))
+import learned_preferences as lp
 config = json.load(open(os.path.expanduser('~/social-autoposter/config.json')))
 projects = config.get('projects', [])
+# learned_preferences is a SINGLE install-wide block since 2026-07-08 (see
+# scripts/learned_preferences.py), not one per project. Computed once here
+# and stamped onto every emitted project dict under the same key so the
+# existing DRAFT_DIRECTIVE prose ('The matched project's learned_preferences
+# block in ALL_PROJECTS_JSON is MANDATORY...') keeps working unchanged.
+global_lp = lp.get_global_block(config)
 lane = os.environ.get('S4L_ACTIVE_LANE', '')
 if lane == 'personal_brand':
     # Personal-brand lane is pure organic growth: the drafter must NOT see any
@@ -1773,10 +1781,17 @@ if lane == 'personal_brand':
     persona = next((p for p in projects if p.get('persona') is True), None)
     out = {}
     if persona:
-        out[persona['name']] = {k: v for k, v in persona.items() if k in ALLOWED}
+        entry = {k: v for k, v in persona.items() if k in ALLOWED}
+        entry['learned_preferences'] = global_lp
+        out[persona['name']] = entry
     print(json.dumps(out, indent=2))
 else:
-    print(json.dumps({p['name']: p for p in projects}, indent=2))
+    out = {}
+    for p in projects:
+        entry = dict(p)
+        entry['learned_preferences'] = global_lp
+        out[p['name']] = entry
+    print(json.dumps(out, indent=2))
 " 2>/dev/null || echo "{}")
 
 # Engagement-style picker (2026-05-19): pick ONE assigned style for this
@@ -1808,6 +1823,44 @@ except Exception:
 " 2>/dev/null)
 log "Engagement style assigned: mode=$PICKED_MODE style=${PICKED_STYLE:-(invent)}"
 
+# --- Second engagement style for dual-draft review cards (2026-07-07) -------
+# Two-draft cards: pick a SECOND style ("Style B") this cycle, independent of
+# the first ("Style A" = PICKED_STYLE/PICKED_MODE above). The prep step drafts
+# every fresh candidate once under EACH style and the reviewer picks which
+# posts; see PREP_SCHEMA/PREP_PROMPT below. Bounded retry so the two styles
+# are actually different (a same-name pair defeats the whole point); INVENT
+# mode on either side is accepted immediately since an invented name is
+# definitionally distinct from a pinned one. This is orthogonal to the
+# treatment_v2/control_v2 draft-prompt A/B below (that varies WORDING of the
+# directive for the whole cycle; this varies STYLE per draft slot), so neither
+# experiment disturbs the other.
+STYLE_ASSIGN_FILE_B=$(mktemp -t s4l_twitter_assign_b_XXXXXX.json)
+for _style_b_attempt in 1 2 3; do
+    s4l_pick_style twitter posting "$STYLE_ASSIGN_FILE_B" >/dev/null 2>&1 || true
+    PICKED_STYLE_B=$(python3 -c "
+import json
+try:
+    with open('$STYLE_ASSIGN_FILE_B') as f:
+        d = json.load(f)
+    print(d.get('style') or '')
+except Exception:
+    print('')
+" 2>/dev/null)
+    PICKED_MODE_B=$(python3 -c "
+import json
+try:
+    with open('$STYLE_ASSIGN_FILE_B') as f:
+        d = json.load(f)
+    print(d.get('mode') or 'use')
+except Exception:
+    print('use')
+" 2>/dev/null)
+    if [ "$PICKED_MODE" = "invent" ] || [ "$PICKED_MODE_B" = "invent" ] || [ "$PICKED_STYLE_B" != "$PICKED_STYLE" ]; then
+        break
+    fi
+done
+log "Engagement style B assigned: mode=$PICKED_MODE_B style=${PICKED_STYLE_B:-(invent)}"
+
 # --- Draft-prompt A/B: decouple product pivot (2026-06-29) -------------------
 # Per-CYCLE arm (the prep session drafts the whole batch from ONE prompt, so
 # assignment is at cycle granularity, not per post; the whole batch shares it).
@@ -1882,6 +1935,11 @@ if [ -n "$PICKED_STYLE" ]; then
 else
     TOP_REPORT=$(python3 "$REPO_DIR/scripts/top_performers.py" --platform twitter 2>/dev/null || echo "(top performers report unavailable)")
 fi
+if [ -n "$PICKED_STYLE_B" ]; then
+    TOP_REPORT_B=$(python3 "$REPO_DIR/scripts/top_performers.py" --platform twitter --style "$PICKED_STYLE_B" 2>/dev/null || echo "(top performers report unavailable)")
+else
+    TOP_REPORT_B=$(python3 "$REPO_DIR/scripts/top_performers.py" --platform twitter 2>/dev/null || echo "(top performers report unavailable)")
+fi
 
 # --- Generation trace -------------------------------------------------------
 # Snapshot the few-shot context this cycle will feed to Claude — top_performers
@@ -1924,9 +1982,26 @@ else
 fi
 
 STYLES_BLOCK=$(s4l_render_style_block "$STYLE_ASSIGN_FILE" twitter posting)
-# Style assignment file is the same one we picked above; styles.sh already sourced.
+# Style B block: the assigned-style portion only (description/example/note/
+# length/grounding rule), NOT the full s4l_render_style_block — content rules,
+# anti-patterns, and the voice-relationship rule are platform/project-wide, not
+# per-style, and are already embedded once via STYLES_BLOCK above. Calling
+# get_assigned_style_prompt() directly here avoids repeating those shared
+# sections a second time and keeps skill/styles.sh untouched.
+STYLES_BLOCK_B=$(python3 -c "
+import json, sys
+sys.path.insert(0, '$REPO_DIR/scripts')
+from engagement_styles import get_assigned_style_prompt
+try:
+    with open('$STYLE_ASSIGN_FILE_B', 'r') as f:
+        assignment = json.load(f)
+    print(get_assigned_style_prompt('twitter', assignment, context='posting'))
+except Exception:
+    print('(style module unavailable)')
+" 2>/dev/null || echo "(style module unavailable)")
+# Style assignment files are the ones we picked above; styles.sh already sourced.
 # Cleanup at cycle end (best effort).
-trap 'rm -f "$STYLE_ASSIGN_FILE" 2>/dev/null || true' EXIT
+trap 'rm -f "$STYLE_ASSIGN_FILE" "$STYLE_ASSIGN_FILE_B" 2>/dev/null || true' EXIT
 
 # Phase 2b is split into three sub-phases so the twitter-browser lock is only
 # held during actual browser work. The killer in the old single-session flow
@@ -2009,22 +2084,26 @@ log "Phase 2b-prep: Claude reading threads and drafting replies (no post cap)...
 CLAUDE_SESSION_ID="$(uuidgen | tr 'A-Z' 'a-z')"
 export CLAUDE_SESSION_ID
 
-# PREP_SCHEMA — strict JSON schema for the prep envelope. Includes
-# optional `new_style` per candidate (an inner object) that the model
-# MUST populate when it chooses a brand-new engagement style name (i.e.
-# the picker set mode=invent and the model invented a snake_case name).
-# Fields mirror engagement_styles.py::_REQUIRED_NEW_STYLE_FIELDS so the
-# downstream validate_or_register call accepts the block without a
-# second schema layer.
-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"},"new_style":{"type":["object","null"],"properties":{"description":{"type":"string"},"example":{"type":"string"},"why_existing_didnt_fit":{"type":"string"},"note":{"type":"string"},"target_chars":{"type":"integer"}}},"language":{"type":"string"},"reply_text_en":{"type":"string"},"thread_text_en":{"type":"string"},"has_landing_pages":{"type":"boolean"},"link_keyword":{"type":"string"},"link_slug":{"type":"string"},"search_topic":{"type":"string"}},"required":["candidate_id","candidate_url","matched_project","reply_text","engagement_style","language","has_landing_pages","search_topic"]}},"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_SCHEMA — strict JSON schema for the prep envelope. Two-draft cards
+# (2026-07-07, no-recommendation redesign 2026-07-08): each candidate
+# carries draft_a_* / draft_b_* (one per assigned style) instead of a
+# single reply_text/engagement_style, plus is_reused_draft. The model does
+# NOT pick a favorite; Draft A is always the default shown/selected on the
+# review card, the reviewer's own click chooses otherwise. Includes optional
+# `draft_a_new_style` / `draft_b_new_style` per candidate (inner objects) that
+# the model MUST populate when that slot's picker set mode=invent and it
+# invented a snake_case name. Fields mirror
+# engagement_styles.py::_REQUIRED_NEW_STYLE_FIELDS so the downstream
+# validate_or_register call accepts the block without a second schema layer.
+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"},"is_reused_draft":{"type":"boolean"},"draft_a_text":{"type":"string"},"draft_a_style":{"type":"string"},"draft_a_new_style":{"type":["object","null"],"properties":{"description":{"type":"string"},"example":{"type":"string"},"why_existing_didnt_fit":{"type":"string"},"note":{"type":"string"},"target_chars":{"type":"integer"}}},"draft_a_text_en":{"type":["string","null"]},"draft_b_text":{"type":["string","null"]},"draft_b_style":{"type":["string","null"]},"draft_b_new_style":{"type":["object","null"],"properties":{"description":{"type":"string"},"example":{"type":"string"},"why_existing_didnt_fit":{"type":"string"},"note":{"type":"string"},"target_chars":{"type":"integer"}}},"draft_b_text_en":{"type":["string","null"]},"language":{"type":"string"},"thread_text_en":{"type":"string"},"has_landing_pages":{"type":"boolean"},"link_keyword":{"type":"string"},"link_slug":{"type":"string"},"search_topic":{"type":"string"}},"required":["candidate_id","candidate_url","matched_project","is_reused_draft","draft_a_text","draft_a_style","draft_b_text","draft_b_style","language","has_landing_pages","search_topic"]}},"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_PROMPT="${TW_ENGINE_PREFIX}You are the Social Autoposter prep step.
 
 Your ONLY job in THIS session:
   1. Read each candidate's thread context from the PRE-SCORED CANDIDATES block below (each entry's 'Text:' field is the parent tweet). You have WebSearch and WebFetch available: use them ONLY when a thread hinges on a current fact, a name, a release, or a claim you are not sure about, so your reply is specific and correct instead of vague. You do NOT have the Twitter/X browser this session — never fetch, navigate, or open a tweet/x.com URL, and never try to load the thread itself; the thread text you need is already inlined below. Most replies need no search at all; reach for it only when it materially improves the reply.
-  2. Draft a reply for each.
-  3. Persist each fresh draft via log_draft.py.
-  4. Emit a structured plan describing the chosen candidates, the reply text, and (when applicable) the SEO link keyword + slug.
+  2. Draft TWO independent replies for each fresh candidate, one under Draft A's assigned style and one under Draft B's assigned style below. Do not judge or rank them, the reviewer reads both and picks.
+  3. Persist the recommended fresh draft via log_draft.py.
+  4. Emit a structured plan describing the chosen candidates, both draft texts, and (when applicable) the SEO link keyword + slug.
 
 You will NOT post anything. You will NOT generate landing pages. You will NOT call log_post.py. The shell handles all of that AFTER your session ends, with the browser lock released for the long landing-page build.
 
@@ -2041,11 +2120,16 @@ $CORPUS_BLOCK
 Each candidate has a 'Project match' field. Use that project unless the thread content clearly better fits another project.
 All project configs: $ALL_PROJECTS_JSON
 
-## FEEDBACK FROM PAST PERFORMANCE:
+## DRAFT A: assigned style + feedback from past performance
 $TOP_REPORT
 
 $STYLES_BLOCK
 
+## DRAFT B: assigned style + feedback from past performance
+$TOP_REPORT_B
+
+$STYLES_BLOCK_B
+
 ## WORKFLOW
 There is NO cap on how many candidates you may pick this cycle. Pick EVERY candidate whose thread is genuinely on-brand and worth a substantive reply. Skip a candidate ONLY when its thread is off-topic for the matched project, toxic / hateful, low-quality / spam, an audience mismatch, or a near-duplicate of something already replied to. Do NOT cap, quota, or balance picks by project: if the strongest candidates this cycle all belong to one project, pick all of them. Project routing matters; project diversification does not. Never force a weak entry just to add volume, and never drop a strong on-brand entry just to limit volume.
 
@@ -2053,14 +2137,18 @@ For each chosen candidate:
 1. Read the candidate's parent tweet from its 'Text:' field in the PRE-SCORED CANDIDATES block above.
 2. Understand the context from that inlined text (the thread text is already in this prompt; you do NOT have the Twitter browser, but you MAY use WebSearch/WebFetch for external facts when a thread needs them to be answered well).
 3. DRAFT HANDLING (existing vs fresh):
-   - If the candidate block shows an EXISTING DRAFT line AND draft age < 30 minutes, REUSE the draft text verbatim. Set engagement_style to the existing style. Do NOT call log_draft.py; do NOT redraft. Reason: prior cycle paid the LLM cost.
-   - $DRAFT_DIRECTIVE
-3a. PERSIST FRESH DRAFTS (skip for reused drafts):
-     python3 $REPO_DIR/scripts/log_draft.py --candidate-id CANDIDATE_ID --text 'YOUR_REPLY_TEXT' --style STYLE --assigned-style '$PICKED_STYLE' --assigned-mode '$PICKED_MODE'
+   - If the candidate block shows an EXISTING DRAFT line AND draft age < 30 minutes, REUSE the draft text verbatim as draft_a_text/draft_a_style (set is_reused_draft=true, draft_b_text=null, draft_b_style=null). Do NOT call log_draft.py; do NOT redraft; do NOT write a second variant, prior cycle already paid the LLM cost for the one draft you have.
+   - Otherwise (fresh candidate, is_reused_draft=false): write TWO independent drafts. Do NOT judge, rank, or pick a favorite between them, both are shown to the reviewer, who decides.
+     - draft_a_text: follow the DRAFT A style block above (its own description/example/note/length limit).
+     - draft_b_text: follow the DRAFT B style block above, written INDEPENDENTLY from scratch as if draft_a_text did not exist. Do NOT lightly reword draft_a_text into draft_b_text, they must diverge in length and rhetorical move because they follow different style templates, not just differ in phrasing. If you notice draft_b_text ending up as a paraphrase of draft_a_text, stop and rewrite it from Style B's own example instead.
+   - $DRAFT_DIRECTIVE (applies to both drafts on fresh candidates; each still obeys its OWN style's length limit, not a shared one).
+3a. PERSIST DRAFT A (skip entirely for reused drafts):
+     python3 $REPO_DIR/scripts/log_draft.py --candidate-id CANDIDATE_ID --text 'DRAFT_A_TEXT' --style DRAFT_A_STYLE --assigned-style '$PICKED_STYLE' --assigned-mode '$PICKED_MODE'
+   Always persist draft_a_text/draft_a_style here (Draft A is the single-draft representative used if a near-immediate next cycle reuses this candidate's draft verbatim per step 3 above); never draft_b.
    The --assigned-style / --assigned-mode flags carry the orchestrator's picker output (this cycle: mode=$PICKED_MODE style='${PICKED_STYLE:-(invent)}') into the candidate row so the post pipeline can coerce drift and register invented styles. Pass them VERBATIM as shown.
-   If you are inventing a brand-new style this cycle (i.e. \$PICKED_MODE=invent and your STYLE is a new snake_case name not in the style block above), ALSO pass:
+   If Draft A used an invented style (i.e. mode is invent and its STYLE is a new snake_case name not in the Draft A style block), ALSO pass:
      --new-style '{\"description\":\"...\",\"example\":\"...\",\"why_existing_didnt_fit\":\"...\"}'
-   with the same description/example/why_existing_didnt_fit you put in the 'new_style' field of your output JSON for this candidate.
+   with the same description/example/why_existing_didnt_fit you put in draft_a_new_style in your output JSON for this candidate.
    Failure here is non-fatal, log a warning and continue.
 4. EMIT one entry in the structured 'candidates' array with these fields:
    - candidate_id (int): from the candidate block
@@ -2068,11 +2156,16 @@ For each chosen candidate:
    - thread_author (string): the @handle (no leading @)
    - thread_text (string): the parent tweet's text, condensed to <=500 chars if needed
    - matched_project (string): the project name to attribute this post to
-   - reply_text (string): the FINAL reply text WITHOUT any URL appended (the shell appends the URL later). 250 chars is the hard ceiling (leaves room for a 23-char t.co link inside the 280-char cap) — stay well under it, not up to it.
-   - engagement_style (string): style name applied (or 'reused' for an unchanged stale draft). In USE mode ($PICKED_MODE=use) this MUST be the assigned style name '${PICKED_STYLE}' verbatim; the orchestrator silently coerces drift back. In INVENT mode ($PICKED_MODE=invent) this MUST be a NEW snake_case style name not in the curated style block.
-   - new_style (object, REQUIRED iff INVENT mode produced a new name; OMIT or set null otherwise): {description (string), example (string), why_existing_didnt_fit (string), note (string, optional), target_chars (integer, REQUIRED)}. Same shape you passed to --new-style in step 3a. The post pipeline reads this and POSTs to /api/v1/engagement-styles/registry so the new style lands in engagement_styles_registry alongside Reddit/GitHub/Moltbook inventions. target_chars is the comment length THIS new style wins at, in characters. IMPORTANT: the example you write must be EXACTLY that length — the example IS the canonical length reference, and the hard ceiling is target_chars × 1.5. Write the example first, count its characters, then set target_chars to that count. Bias SHORT: one-liner style ~45, story-arc style up to ~180, never above 220.
+   - is_reused_draft (bool, REQUIRED): true iff you reused an existing draft verbatim per step 3 above, false for a freshly-drafted candidate.
+   - draft_a_text (string, REQUIRED): the FINAL Draft A reply text WITHOUT any URL appended (the shell appends the URL later). 250 chars is the hard ceiling (leaves room for a 23-char t.co link inside the 280-char cap) — stay well under it, not up to it. On a reused candidate this IS the reused text.
+   - draft_a_style (string, REQUIRED): style name applied to draft_a_text (or the reused candidate's existing style). In USE mode ($PICKED_MODE=use) this MUST be the Draft A assigned style name '${PICKED_STYLE}' verbatim; the orchestrator silently coerces drift back. In INVENT mode ($PICKED_MODE=invent) this MUST be a NEW snake_case style name not in the Draft A style block.
+   - draft_a_new_style (object, REQUIRED iff Draft A's INVENT mode produced a new name; OMIT or set null otherwise): {description (string), example (string), why_existing_didnt_fit (string), note (string, optional), target_chars (integer, REQUIRED)}. target_chars is the comment length THIS new style wins at, in characters; the example you write must be EXACTLY that length, write the example first, count its characters, then set target_chars to that count. Bias SHORT: one-liner style ~45, story-arc style up to ~180, never above 220.
+   - draft_a_text_en (string, REQUIRED when language != 'en'; null when language == 'en'): a faithful English translation of draft_a_text. Display-only: the review card shows it so the operator can understand a non-English draft; it is NEVER posted. Translate meaning, not word-by-word; no added commentary.
+   - draft_b_text (string, REQUIRED when is_reused_draft=false; null when is_reused_draft=true): the FINAL Draft B reply text, same rules as draft_a_text, written under the DRAFT B style block instead.
+   - draft_b_style (string, REQUIRED when is_reused_draft=false; null when is_reused_draft=true): style name applied to draft_b_text, same rules as draft_a_style but against the Draft B assignment (USE mode style '${PICKED_STYLE_B}', mode $PICKED_MODE_B).
+   - draft_b_new_style (object, REQUIRED iff Draft B's INVENT mode produced a new name; OMIT or set null otherwise): same shape as draft_a_new_style.
+   - draft_b_text_en (string, REQUIRED when is_reused_draft=false AND language != 'en'; null otherwise): faithful English translation of draft_b_text, same display-only rules as draft_a_text_en.
    - language (string): ISO 639-1 code (en, ja, zh, es, ...)
-   - reply_text_en (string, REQUIRED when language != 'en'; OMIT when language == 'en'): a faithful English translation of reply_text. Display-only: the review card shows it so the operator can understand a non-English draft; it is NEVER posted. Translate meaning, not word-by-word; no added commentary.
    - thread_text_en (string, REQUIRED when language != 'en'; OMIT when language == 'en'): a faithful English translation of thread_text (same <=500 char condensation). Display-only, never posted.
    - has_landing_pages (bool): true iff the matched project has BOTH landing_pages.repo AND landing_pages.base_url set in config.json. Otherwise false.
    - link_keyword (string, REQUIRED when has_landing_pages=true; OMIT otherwise): a SHORT 3-6 word phrase that captures the ESSENCE OF YOUR REPLY (not just the thread topic). Think: what would a reader search to find a useful page about what you just said?
@@ -2267,6 +2360,47 @@ except Exception as _e:
     print(f'prep: experiments stamp failed: {_e}', file=sys.stderr)
 for _c in candidates:
     _c['experiments'] = dict(_exps)
+    # Two-draft cards (2026-07-07; no-recommendation redesign 2026-07-08):
+    # mirror Draft A onto the canonical single-draft fields (reply_text/
+    # engagement_style/new_style/reply_text_en/assigned_style/assigned_mode)
+    # so every existing downstream consumer (tail-link baking, dedup,
+    # translations, posting, drift coercion) keeps working unchanged by
+    # default. Draft A is always the default, the model is never asked to
+    # pick a favorite, only the reviewer's own click on the card overrides
+    # it. Additionally attach a 'drafts' array for the review card to offer
+    # a switch; reused/stale candidates (single draft, no draft_b) don't get
+    # a drafts array, so the card falls back to today's single-draft UI.
+    _rec = 'a'
+    _is_reused = bool(_c.get('is_reused_draft'))
+    _rec_text = _c.get(f'draft_{_rec}_text') or ''
+    _rec_style = _c.get(f'draft_{_rec}_style') or ''
+    _rec_new_style = _c.get(f'draft_{_rec}_new_style')
+    _rec_text_en = _c.get(f'draft_{_rec}_text_en')
+    _c['reply_text'] = _rec_text
+    _c['engagement_style'] = _rec_style
+    _c['new_style'] = _rec_new_style
+    if _rec_text_en:
+        _c['reply_text_en'] = _rec_text_en
+    _c['assigned_style'] = ('$PICKED_STYLE' or None) if _rec == 'a' else ('$PICKED_STYLE_B' or None)
+    _c['assigned_mode'] = ('$PICKED_MODE' or 'use') if _rec == 'a' else ('$PICKED_MODE_B' or 'use')
+    _draft_b_text = _c.get('draft_b_text')
+    if not _is_reused and _draft_b_text:
+        _c['drafts'] = [
+            {
+                'variant': 'a', 'text': _c.get('draft_a_text') or '',
+                'style': _c.get('draft_a_style') or '',
+                'text_en': _c.get('draft_a_text_en'),
+                'assigned_style': '$PICKED_STYLE' or None,
+                'assigned_mode': '$PICKED_MODE' or 'use',
+            },
+            {
+                'variant': 'b', 'text': _draft_b_text,
+                'style': _c.get('draft_b_style') or '',
+                'text_en': _c.get('draft_b_text_en'),
+                'assigned_style': '$PICKED_STYLE_B' or None,
+                'assigned_mode': '$PICKED_MODE_B' or 'use',
+            },
+        ]
 json.dump({'candidates': candidates,
            'session_id': '$CLAUDE_SESSION_ID',
            'assigned_style': '$PICKED_STYLE' or None,
@@ -2367,6 +2501,44 @@ if [ "$GEN_EXIT" -ne 0 ]; then
     log "WARN: twitter_gen_links.py exited $GEN_EXIT, continuing with whatever links it set (per-candidate fallback to plain project URL on gen failure)."
 fi
 
+# Two-draft cards: twitter_gen_links.py (above, locked/untouched) bakes the
+# tail link into reply_text in place for Draft A only (the canonical default),
+# it never looks at the 'drafts' array. Sync the same suffix onto Draft B so
+# switching to it on the review card still posts with the link.
+# link_tail.py's apply_tail_link is documented APPEND-ONLY (concatenates the
+# bridge sentence onto the unmodified reply_text, never rewrites it), so
+# diffing the appended suffix and re-applying it verbatim is safe.
+python3 -c "
+import json
+plan_path = '$PLAN_FILE'
+try:
+    with open(plan_path) as f:
+        plan = json.load(f)
+except Exception as e:
+    print(f'[gen] drafts-link-sync: plan read failed: {e}')
+    raise SystemExit(0)
+changed = 0
+for c in plan.get('candidates') or []:
+    drafts = c.get('drafts')
+    if not isinstance(drafts, list) or len(drafts) != 2:
+        continue
+    rec_idx = 0  # Draft A is always the canonical default reply_text mirrors
+    other_idx = 1
+    rec_draft = drafts[rec_idx]
+    other_draft = drafts[other_idx]
+    baked = (c.get('reply_text') or '')
+    original = (rec_draft.get('text') or '')
+    if baked and original and baked != original and baked.startswith(original):
+        suffix = baked[len(original):]
+        rec_draft['text'] = baked
+        other_draft['text'] = (other_draft.get('text') or '') + suffix
+        changed += 1
+if changed:
+    with open(plan_path, 'w') as f:
+        json.dump(plan, f, indent=2)
+print(f'[gen] drafts-link-sync: synced tail link onto the non-recommended draft for {changed} candidate(s)')
+" 2>&1 | tee -a "$LOG_FILE"
+
 # --- DRAFT_ONLY gate: stop after drafting for human review (MCP manual mode) -
 # When DRAFT_ONLY=1 the cycle runs scan -> score -> draft -> link-gen, leaves the
 # fully-baked plan (links already stamped into reply_text) at $PLAN_FILE, and