@sellable/mcp 0.1.247 → 0.1.249

package/README.md

@@ -16,9 +16,10 @@ Each message gets 5+ minutes of Claude attention with deep research - no other t
 
 ### Prompt Source Of Truth
 
-There are three public Sellable entrypoints shared across hosts:
+There are four public Sellable entrypoints shared across hosts:
 
 - `sellable:create-campaign`
+- `sellable:content`
 - `sellable:interview`
 - `sellable:create-post`
 
@@ -33,6 +34,12 @@ from:
 
 - `mcp/sellable/skills/interview/SKILL.md`
 
+The content public wrapper captures transcripts and rough notes, clusters
+recurring ideas, extracts story/proof/question cards, and develops post seeds
+before drafting from:
+
+- `mcp/sellable/skills/content/SKILL.md`
+
 The create-post public wrapper captures raw ideas, loads voice internally,
 researches hooks plus market beliefs, develops a premise with story/tension and
 reader value, validates proof/AI tells, and saves content artifacts under
@@ -137,9 +144,9 @@ The installer does the full local setup:
   `mcp__sellable__*` tools into skill sessions
 
 After the installer passes, fully quit and reopen Codex Desktop. Start a new
-thread and select `Sellable Create Campaign`, `Sellable Identity Interview`, or
-`Sellable Create Post`; or invoke `$sellable:create-campaign`,
-`$sellable:interview`, or `$sellable:create-post`. If the app still says
+thread and select `Sellable Create Campaign`, `Sellable Content`,
+`Sellable Identity Interview`, or `Sellable Create Post`; or invoke `$sellable:create-campaign`,
+`$sellable:content`, `$sellable:interview`, or `$sellable:create-post`. If the app still says
 `mcp__sellable__*` tools are missing after the installer passes, check that
 `~/.codex/config.toml` contains both `[marketplaces.sellable]` and
 `[plugins."sellable@sellable"]`.
@@ -151,16 +158,20 @@ the Sellable MCP tools for Codex Desktop.
 Use these names consistently:
 
 - Claude Code command: `/sellable:create-campaign`
+- Claude Code command: `/sellable:content`
 - Claude Code command: `/sellable:interview`
 - Claude Code command: `/sellable:create-post`
 - Codex command: `$sellable:create-campaign`
+- Codex command: `$sellable:content`
 - Codex command: `$sellable:interview`
 - Codex command: `$sellable:create-post`
 - Codex Desktop plugin: `sellable@sellable`
 - Codex visible skill: `Sellable Create Campaign`
+- Codex visible skill: `Sellable Content`
 - Codex visible skill: `Sellable Identity Interview`
 - Codex visible skill: `Sellable Create Post`
 - Codex skill frontmatter name: `create-campaign`
+- Codex skill frontmatter name: `content`
 - Codex skill frontmatter name: `interview`
 - Codex skill frontmatter name: `create-post`
 - MCP server name: `sellable`
@@ -170,8 +181,10 @@ Never tell users to run `/sellable:create-campaign-v2`,
 `$sellable:create-campaign-v2`, `$sellable:load-voice`, or
 `$sellable:sellable:create-campaign`. `create-campaign-v2` is an internal MCP
 subskill loaded by `get_subskill_prompt({ subskillName: "create-campaign-v2" })`.
-`load-voice` is an internal direct MCP utility; `create-post` loads voice
-silently for post drafting.
+`load-voice` is an internal direct MCP utility. `content` is the preferred
+entrypoint for adding transcripts, recurring ideas, and post seeds; `create-post`
+remains a supported direct drafting shortcut and loads voice silently for post
+drafting.
 
 ## Structured Question Parity
 

package/dist/tools/campaigns.js

@@ -49,31 +49,16 @@ function getCampaignBuilderWatchModeFromUrl(watchUrl) {
 export function getCampaignBuilderWatchModeDriverLabel(mode = getCampaignBuilderWatchModeParam()) {
     return mode === "codex" ? "Codex" : "Claude Code";
 }
-function buildWatchHandoffHeadlineBox(headline) {
-    const width = 52;
-    const visibleHeadline = headline.length > width ? headline.slice(0, width) : headline;
-    const padding = Math.max(width - visibleHeadline.length, 0);
-    const left = Math.floor(padding / 2);
-    const right = padding - left;
-    const border = `+${"-".repeat(width + 2)}+`;
-    return [
-        "```",
-        border,
-        `| ${" ".repeat(left)}${visibleHeadline}${" ".repeat(right)} |`,
-        border,
-        "```",
-    ].join("\n");
-}
 export function buildCampaignWatchHandoffMarkdown(watchUrl, mode = getCampaignBuilderWatchModeFromUrl(watchUrl) ??
     getCampaignBuilderWatchModeParam()) {
     const driverLabel = getCampaignBuilderWatchModeDriverLabel(mode);
     const headline = `WATCH ${driverLabel.toUpperCase()} BUILD THE CAMPAIGN LIVE`;
     return [
-        buildWatchHandoffHeadlineBox(headline),
-        "",
-        `[Open live campaign builder](${watchUrl})`,
-        "",
-        "Keep this chat open. I'll ask approval questions here before making decisions that need your judgment.",
+        `> **${headline}**`,
+        ">",
+        `> [Open live campaign builder](${watchUrl})`,
+        ">",
+        "> Keep this chat open. I'll ask approval questions here before making decisions that need your judgment.",
     ].join("\n");
 }
 function buildCampaignBuilderWatchPath(campaignId) {

package/dist/tools/prompts.d.ts

@@ -135,7 +135,7 @@ export interface PostFindLeadsScoutRegistryResponse {
 }
 export declare const DEFAULT_SUBSKILL_PROMPT_CHUNK_CHARS = 48000;
 export declare const MAX_SUBSKILL_PROMPT_CHUNK_CHARS = 48000;
-export declare const ALLOWED_SUBSKILL_PROMPT_NAMES: readonly ["building-gtm-tables", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
+export declare const ALLOWED_SUBSKILL_PROMPT_NAMES: readonly ["building-gtm-tables", "content", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
 export declare const promptToolDefinitions: ({
     name: string;
     description: string;
@@ -179,7 +179,7 @@ export declare const promptToolDefinitions: ({
         properties: {
             subskillName: {
                 type: string;
-                enum: readonly ["building-gtm-tables", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
+                enum: readonly ["building-gtm-tables", "content", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
                 description: string;
             };
             offset: {
@@ -214,7 +214,7 @@ export declare const promptToolDefinitions: ({
         properties: {
             subskillName: {
                 type: string;
-                enum: readonly ["building-gtm-tables", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
+                enum: readonly ["building-gtm-tables", "content", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
                 description: string;
             };
             assetPath: {

package/dist/tools/prompts.js

@@ -18,6 +18,7 @@ const DEPRECATED_SUBSKILL_PROMPT_REPLACEMENTS = {
 };
 export const ALLOWED_SUBSKILL_PROMPT_NAMES = [
     "building-gtm-tables",
+    "content",
     "create-campaign",
     "create-campaign-brief",
     "create-campaign-v2",

package/dist/tools/registry.d.ts

@@ -225,7 +225,7 @@ export declare const allTools: ({
         properties: {
             subskillName: {
                 type: string;
-                enum: readonly ["building-gtm-tables", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
+                enum: readonly ["building-gtm-tables", "content", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
                 description: string;
             };
             offset: {
@@ -260,7 +260,7 @@ export declare const allTools: ({
         properties: {
             subskillName: {
                 type: string;
-                enum: readonly ["building-gtm-tables", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
+                enum: readonly ["building-gtm-tables", "content", "create-campaign", "create-campaign-brief", "create-campaign-v2", "create-campaign-v2-tail", "create-campaign-v2-validation", "create-post", "create-rubric", "engage", "enrich-prospects", "find-leads", "generate-messages", "interview", "load-voice", "research", "research-prospect", "research-sender", "workflow-sequences"];
                 description: string;
             };
             assetPath: {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@sellable/mcp",
-  "version": "0.1.247",
+  "version": "0.1.249",
   "type": "module",
   "description": "Sellable MCP server for Claude Code and Codex campaign workflows",
   "main": "dist/index.js",

package/skills/content/SKILL.md

@@ -0,0 +1,213 @@
+---
+name: content
+description: Capture, cluster, and develop transcript-driven content ideas before drafting LinkedIn posts through create-post.
+visibility: public
+---
+
+# Content
+
+<command_model>
+`content` is the public umbrella command for content raw material.
+
+Use it when the user wants to add ideas, transcripts, voice notes, rough notes,
+recurring themes, story fragments, proof snippets, open questions, or post seeds.
+It owns capture, clustering, ideation, and premise development.
+
+It does not own a separate drafting engine. When the user asks for an actual
+LinkedIn post draft, load and follow `create-post` exactly:
+
+`mcp__sellable__get_subskill_prompt({ subskillName: "create-post" })`
+
+`create-post` remains the post engine for hook research, market-belief research,
+drafting, validation receipts, draft iterations, publishing records, and final
+save behavior.
+
+`$sellable:interview` remains the foundation-memory command for core values, company
+truth, proof hygiene, durable stories, reusable answers, and anti-AI writing
+rules. `content` reads that foundation through
+`mcp__sellable__get_engage_memory`; it does not duplicate it.
+</command_model>
+
+<role>
+You are the Sellable content memory librarian and post ideation partner.
+
+Your job is to preserve the user's raw source material, identify recurring
+ideas across time, cluster related notes without flattening nuance, extract
+story/proof/question cards, and help develop the meat of a post before drafting.
+
+Hard fail patterns:
+
+- inventing stories, claims, metrics, customers, or proof
+- creating a second post drafting workflow instead of delegating to `create-post`
+- treating every raw note as ready to publish
+- losing the user's raw phrasing
+- splitting one evolving idea into many duplicate clusters
+- merging unrelated ideas only because they share generic words
+- moving core identity/proof/story memory out of `~/.sellable/configs/core/**`
+- calling campaign, outbound, or message-generation tools
+</role>
+
+<tools>
+Use these MCP tools when available:
+
+- `mcp__sellable__get_auth_status`
+- `mcp__sellable__get_engage_memory`
+- `mcp__sellable__get_subskill_prompt`
+- `mcp__sellable__get_subskill_asset`
+- `mcp__sellable__search_subskill_prompts`
+- `mcp__sellable__capture_post_idea`
+- `mcp__sellable__get_post_idea`
+- `mcp__sellable__list_post_ideas`
+- `mcp__sellable__save_hook_research`
+- `mcp__sellable__save_post_draft`
+- `mcp__sellable__update_post_draft`
+- `mcp__sellable__list_post_draft_iterations`
+- `mcp__sellable__get_post_draft`
+- `mcp__sellable__mark_post_published`
+- `mcp__sellable__get_published_post`
+- `mcp__sellable__update_published_post_metrics`
+- `mcp__sellable__list_published_posts`
+- `mcp__sellable__search_engagement_posts`
+- `mcp__sellable__fetch_linkedin_posts`
+- `mcp__sellable__fetch_linkedin_profile`
+- `mcp__sellable__record_engage_proven_search`
+- `mcp__sellable__upsert_engage_tracked_person`
+
+Do not call outbound/campaign tools from this skill. Do not call
+`generate-messages`, campaign lead tools, comment drafting tools, or campaign
+approval tools.
+</tools>
+
+<memory_contract>
+Always start by loading unified memory:
+
+1. Call `mcp__sellable__get_engage_memory`.
+2. Use the returned foundation memory when present:
+   - `core/about-me.md`
+   - `core/my-company.md`
+   - `core/story-bank.md`
+   - `core/proof-ledger.md`
+   - `core/wins-ledger.md`
+   - `core/answer-bank.md`
+   - `core/context-modes.md`
+   - `core/anti-ai-writing-style.md`
+3. Use content memory when present:
+   - `core/content-memory/INDEX.md`
+   - `core/content-memory/clusters/**`
+   - `core/content-memory/cards/story-cards.md`
+   - `core/content-memory/cards/proof-cards.md`
+   - `core/content-memory/cards/question-cards.md`
+   - `core/content-memory/cards/post-seeds.md`
+
+`get_engage_memory` is the backward-compatible tool name for unified Sellable
+memory. Treat `memory.contentMemory` as the current content-intelligence layer.
+
+Raw post ideas and drafts live under `~/.sellable/content/linkedin/**` through
+the content tools. Evolving content memory lives under
+`~/.sellable/configs/core/content-memory/**`.
+
+If the user explicitly asks to add, log, organize, or cluster source material,
+that is permission to write the smallest relevant content-memory update. If the
+update would promote a private story, unverified proof, or a durable core
+identity/company claim, ask before writing it to core proof/story/answer files.
+</memory_contract>
+
+<modes>
+## Capture Only
+
+Use when the user says to add, log, save, capture, or preserve raw material and
+does not ask for a draft.
+
+1. Call `mcp__sellable__capture_post_idea`.
+2. Preserve the raw source exactly.
+3. Distill only what the user actually said.
+4. Match the source against existing `core/content-memory/**`.
+5. Add it to an existing cluster when it is the same evolving idea.
+6. Create a new cluster only when no existing cluster fits.
+7. Return the saved idea ID, cluster decision, and strongest follow-up questions.
+
+Do not draft in capture-only mode.
+
+## Cluster
+
+Use when the user asks whether repeated mentions should be bunched together,
+how an idea is evolving, or what raw material exists for a theme.
+
+Cluster by underlying claim, tension, story mechanism, proof basis, and intended
+reader value. Do not cluster only by surface keywords.
+
+For each cluster, maintain:
+
+- stable cluster ID and title
+- source keys and source idea IDs
+- current thesis
+- evolution notes
+- reusable story cards
+- proof cards with verification status
+- question cards for what the system should ask next
+- post seeds with maturity: raw, promising, develop, draft-ready
+
+## Ideate
+
+Use when the user asks for post ideas, hidden insights, angles they may be
+missing, or what the transcript library suggests.
+
+Mine `memory.contentMemory`, raw ideas, story cards, proof cards, and question
+cards. Return high-signal post seeds, not finished posts. Each seed should
+include:
+
+- premise
+- why this matters now
+- source material
+- reader tension
+- proof/story available
+- missing question
+- suggested next action
+
+## Develop
+
+Use when a seed is promising but not draft-ready.
+
+Develop the meat of the post before hooks:
+
+- the specific scene, repeated pattern, or lived observation
+- the common belief it challenges
+- the contrarian or under-said truth
+- why the user can credibly say it
+- what the reader gets out of it
+- what proof/story is available
+- what question needs to be answered before drafting
+
+Ask 1-3 focused questions when story, proof, audience, or stakes are missing.
+
+## Draft Handoff
+
+Use when the user asks for a LinkedIn post, draft, hooks, final copy, publishable
+version, or asks to run the post pipeline.
+
+Do not draft from this prompt. Load the canonical post engine and follow it:
+
+1. `mcp__sellable__get_subskill_prompt({ subskillName: "create-post" })`
+2. Continue chunked prompt loads until `hasMore=false`.
+3. Load every `create-post` required asset with
+   `mcp__sellable__get_subskill_asset`.
+4. Follow `create-post` exactly for hook research, market-belief research,
+   premise cards, drafting, validation, iteration, and saving.
+</modes>
+
+<response_shape>
+For capture, cluster, ideation, or development work, return:
+
+```text
+status: captured | clustered | ideated | needs_questions | handed_to_create_post | blocked
+idea_id: <id or none>
+cluster_id: <id or none>
+cluster_decision: existing | new | none
+source_keys: <compact list>
+post_seeds: <compact list or none>
+questions: <1-3 focused questions or none>
+next_step: <best next action>
+```
+
+For drafts, use the `create-post` response shape, not a content-specific shape.
+</response_shape>

package/skills/create-campaign/SKILL.md

@@ -409,23 +409,19 @@ automation just because a watch link exists. If `create_campaign` returns
 approval question. It will use the URL mode to say Codex or Claude Code:
 
 ````markdown
-```
-+------------------------------------------------------+
-|         WATCH CODEX BUILD THE CAMPAIGN LIVE          |
-+------------------------------------------------------+
-```
-
-[Open live campaign builder]({watchUrl})
-
-Keep this chat open. I'll ask approval questions here before making decisions
-that need your judgment.
+> **WATCH CODEX BUILD THE CAMPAIGN LIVE**
+>
+> [Open live campaign builder]({watchUrl})
+>
+> Keep this chat open. I'll ask approval questions here before making decisions
+> that need your judgment.
 ````
 
-The ASCII banner is intentional because it is unmissable in both Codex and plain
-terminals. Only the banner should be fenced/preformatted; keep the Markdown link
-and chat note outside the fence so rendered chat clients still expose a normal
-click target. Do not add a code-fence language label, replace the link with a
-shell command, or add a browser-opening instruction.
+The rendered callout is intentional: rendered chat clients highlight the
+headline, link, and note without treating it as a code block, and plain terminals
+still expose the tokenized URL inside the Markdown target. Do not wrap this CTA
+in a fenced code block, replace it with a shell command, or add a
+browser-opening instruction.
 
 The watch link should auto-login through the token in the URL. If the user says
 the link lands on auth, 404, permission, blank, or a visible error state, recover

package/skills/create-campaign-v2/SOUL.md

@@ -341,16 +341,12 @@ approve/revise recommendation, not fallback samples, concerns, or a QA receipt.
 Be explicit about what has and has not happened:
 
 ````markdown
-```
-+------------------------------------------------------+
-|         WATCH CODEX BUILD THE CAMPAIGN LIVE          |
-+------------------------------------------------------+
-```
-
-[Open live campaign builder]({watchUrl})
-
-Keep this chat open. I'll ask approval questions here before making decisions
-that need your judgment.
+> **WATCH CODEX BUILD THE CAMPAIGN LIVE**
+>
+> [Open live campaign builder]({watchUrl})
+>
+> Keep this chat open. I'll ask approval questions here before making decisions
+> that need your judgment.
 ````
 
 Print that watch-link handoff only once before the brief approval question, and

package/skills/create-campaign-v2/references/watch-link-handoff.md

@@ -39,16 +39,12 @@ useful brief instead of an empty campaign.
 Example skeleton:
 
 ````markdown
-```
-+------------------------------------------------------+
-|         WATCH CODEX BUILD THE CAMPAIGN LIVE          |
-+------------------------------------------------------+
-```
-
-[Open live campaign builder]({watchUrl})
-
-Keep this chat open. I'll ask approval questions here before making decisions
-that need your judgment.
+> **WATCH CODEX BUILD THE CAMPAIGN LIVE**
+>
+> [Open live campaign builder]({watchUrl})
+>
+> Keep this chat open. I'll ask approval questions here before making decisions
+> that need your judgment.
 
 Now ask the brief approval question. Do not show the approve/revise question
 panel until the full brief content and this exact watch-link handoff have

package/skills/create-post/SKILL.md

@@ -1,7 +1,7 @@
 ---
 name: create-post
 description: Capture rough LinkedIn post ideas, preserve the raw source, develop a valuable premise, research currently working hooks, and save validated drafts in the user's voice.
-visibility: internal
+visibility: public
 ---
 
 # Create Post

package/skills/research/config.json

@@ -0,0 +1,9 @@
+{
+  "parallelMode": "wide",
+  "agentCount": 6,
+  "maxToolCallsPerAgent": 2,
+  "senderMaxAgents": 2,
+  "senderMaxToolCallsPerAgent": 3,
+  "progressMode": true,
+  "debugMode": true
+}