@sellable/install 0.1.179 → 0.1.181

package/README.md

@@ -23,8 +23,8 @@ agent command for launching a campaign:
 sellable create
 ```
 
-Campaign creation itself runs inside Claude Code or Codex, where the Sellable
-MCP tools and approval flow are available.
+Campaign creation, post drafting, and identity interviews run inside Claude Code
+or Codex, where the Sellable MCP tools and approval flows are available.
 
 Install is auth-free by default. If you do not pass a token, the agent handles
 Sellable sign-in on the first campaign run with a magic-link handoff.
@@ -70,19 +70,21 @@ Use the same public entrypoints in both hosts:
 
 - Claude Code: `/sellable:create-campaign`
 - Claude Code: `/sellable:interview`
-- Claude Code: `/sellable:load-voice`
+- Claude Code: `/sellable:create-post`
 - Codex: `$sellable:create-campaign`
 - Codex: `$sellable:interview`
-- Codex: `$sellable:load-voice`
+- Codex: `$sellable:create-post`
 - Codex Desktop plugin: `sellable@sellable`
 - Codex visible skill: `Sellable Create Campaign`
 - Codex visible skill: `Sellable Identity Interview`
-- Codex visible skill: `Sellable Load Voice`
+- Codex visible skill: `Sellable Create Post`
 - Internal MCP workflow prompt: `create-campaign-v2`
 
 Do not ask users to run `/sellable:create-campaign-v2`,
-`$sellable:create-campaign-v2`, or `$sellable:sellable:create-campaign`.
-`create-campaign-v2` is loaded internally by the skill.
+`$sellable:create-campaign-v2`, `$sellable:load-voice`, or
+`$sellable:sellable:create-campaign`. `create-campaign-v2` is loaded internally
+by the campaign skill. `create-post` loads voice internally, so there is no
+separate public voice-loading command.
 
 ## Structured Questions
 

package/bin/sellable-install.mjs

@@ -31,7 +31,7 @@ function getInstallVersion() {
   }
 }
 
-const CODEX_PLUGIN_VERSION = "0.1.34";
+const CODEX_PLUGIN_VERSION = "0.1.36";
 const CODEX_PLUGIN_COMPAT_VERSIONS = [
   "0.1.8",
   "0.1.9",
@@ -59,6 +59,8 @@ const CODEX_PLUGIN_COMPAT_VERSIONS = [
   "0.1.31",
   "0.1.32",
   "0.1.33",
+  "0.1.34",
+  "0.1.35",
 ];
 const INSTALL_PACKAGE_SPEC =
   process.env.SELLABLE_INSTALL_PACKAGE_SPEC || "@sellable/install@latest";
@@ -149,9 +151,9 @@ Options:
 
 Auth:
   Install is auth-free by default. Sign in happens on the first run of
-  /sellable:create-campaign in Claude Code or Codex, where the agent walks
-  you through signup or sign-in and stores credentials in
-  ~/.sellable/config.json.
+  /sellable:create-campaign or /sellable:create-post in Claude Code or Codex,
+  where the agent walks you through signup or sign-in and stores credentials
+  in ~/.sellable/config.json.
 
   For CI/scripted installs, pass --token + --workspace-id or set
   SELLABLE_TOKEN + SELLABLE_WORKSPACE_ID and the installer will write the
@@ -174,11 +176,15 @@ function printCreateCommandHint() {
   console.log("");
   printAgentBox("Using Claude Code?", "claude", [
     { label: "Campaign", command: "/sellable:create-campaign" },
+    { label: "Post", command: "/sellable:create-post" },
+    { label: "Interview", command: "/sellable:interview" },
   ]);
   console.log("");
   console.log("");
   printAgentBox("Using Codex?", "codex", [
     { label: "Campaign", command: "$sellable:create-campaign" },
+    { label: "Post", command: "$sellable:create-post" },
+    { label: "Interview", command: "$sellable:interview" },
   ]);
   console.log("");
   console.log(`   ${"─".repeat(63)}`);
@@ -839,16 +845,18 @@ automation just because a watch link exists. If \`create_campaign\` returns
 \`watchHandoff.markdown\`, print that exact value once, directly before the brief
 approval question. It will use the URL mode to say Codex or Claude Code:
 
+\`\`\`\`markdown
 \`\`\`text
-╔══════════════════════════════════════════════════════════════╗
-║    OPEN THIS LINK TO WATCH CODEX BUILD THE CAMPAIGN LIVE    ║
-╚══════════════════════════════════════════════════════════════╝
++------------------------------------------------------+
+|         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 Markdown link is intentional: rendered chat clients turn it into a large
 click target, and plain terminals still expose the tokenized URL inside the
@@ -1226,6 +1234,7 @@ allowed-tools:
   - mcp__sellable__get_subskill_prompt
   - mcp__sellable__get_subskill_asset
   - mcp__sellable__search_subskill_prompts
+  - mcp__sellable__capture_post_idea
   - Read
   - Write
   - Edit
@@ -1262,6 +1271,14 @@ Desktop, then start a new thread.
 3. Follow the canonical prompt exactly. Save local memory only where that prompt
    directs, under \`~/.sellable/configs/core/**\` and
    \`~/.sellable/interviews/**\`.
+4. If the interview captures a content-specific LinkedIn post idea, hook/taste
+   correction, draft-first post calibration note, or voice-note transcript meant
+   as a post idea, the canonical prompt may bridge only that item with
+   \`mcp__sellable__capture_post_idea\` under
+   \`~/.sellable/content/linkedin/ideas/**\`. Core identity, proof, stories,
+   answer-bank entries, transcripts, references, and raw archives must stay in
+   \`~/.sellable/configs/core/**\` and \`~/.sellable/interviews/**\`.
+   Core identity, proof, stories, answer-bank entries, transcripts, references, and raw archives must stay in core/interviews.
 
 ## MCP Prompt Fallback
 
@@ -1271,27 +1288,40 @@ then retry \`get_subskill_prompt\`.
 `;
 }
 
-function loadVoiceSkillMd() {
+function createPostSkillMd() {
   return `---
-name: load-voice
-description: Load Sellable voice/company memory for answering questions, writing posts, applications, replies, and reviews in the user's voice.
+name: create-post
+description: Capture rough LinkedIn post ideas, research currently working hooks, and save validated drafts in the user's voice.
 allowed-tools:
+  - mcp__sellable__get_auth_status
   - mcp__sellable__get_subskill_prompt
+  - mcp__sellable__get_subskill_asset
   - mcp__sellable__search_subskill_prompts
-  - Read
-  - Glob
-  - Grep
+  - mcp__sellable__get_engage_memory
+  - 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__get_post_draft
+  - mcp__sellable__mark_post_published
+  - 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
 ---
 
-# Sellable Load Voice
+# Sellable Create Post
 
-Use this as the customer-facing entrypoint for loading Sellable voice/company
-memory before drafting, answering, rewriting, reviewing, or calibrating copy in
-the user's voice.
+Use this as the customer-facing entrypoint for the Sellable \`create-post\`
+workflow. It captures raw LinkedIn post ideas, preserves the user's source
+wording, researches currently working hooks, validates proof/voice/AI tells, and
+saves drafts under \`~/.sellable/content/linkedin/**\`.
 
 ## Bootstrap
 
-MCP prompt access and local read tools are required. Do not run shell commands,
+MCP prompt access is required. Do not inspect repo files, run shell commands,
 use \`npm\`, \`node\`, local harness scripts, or read local prompt files to
 emulate this workflow.
 
@@ -1302,20 +1332,24 @@ Desktop, then start a new thread.
 
 ## Execute Workflow
 
-1. Load the canonical prompt via
-   \`mcp__sellable__get_subskill_prompt({ subskillName: "load-voice" })\`.
+1. Call \`mcp__sellable__get_auth_status({})\` so hook research failures can be
+   distinguished from local content capture.
+2. Load the canonical prompt via
+   \`mcp__sellable__get_subskill_prompt({ subskillName: "create-post" })\`.
    If the response has \`hasMore=true\`, continue with \`nextOffset\` until
    \`hasMore=false\`.
-2. Follow the canonical prompt exactly. Read the relevant
-   \`~/.sellable/configs/**\` memory files it names.
-3. Apply the loaded memory silently to the user's requested writing or answer.
-   If the user only asked to load voice, summarize the active rules briefly and
-   ask what they want drafted, answered, or reviewed.
+3. Load every required create-post reference with
+   \`mcp__sellable__get_subskill_asset({ subskillName: "create-post", assetPath: "references/<file>.md" })\`.
+   If any asset response has \`hasMore=true\`, continue with \`nextOffset\`
+   until \`hasMore=false\`.
+4. Follow the canonical prompt exactly. Use the content tools for ideas,
+   research, drafts, and published records. Do not write repo-local prompt files
+   or append to legacy flat draft files.
 
 ## MCP Prompt Fallback
 
 If exact subskill lookup fails, use
-\`mcp__sellable__search_subskill_prompts({ query: "load-voice", includePublic: true })\`,
+\`mcp__sellable__search_subskill_prompts({ query: "create-post", includePublic: true, includeInternal: true })\`,
 then retry \`get_subskill_prompt\`.
 `;
 }
@@ -1524,10 +1558,10 @@ function codexPluginSkills() {
       skillMd: interviewSkillMd(),
     },
     {
-      dir: "sellable-load-voice",
-      displayName: "Sellable Load Voice",
-      description: "Load voice memory for writing and answers",
-      skillMd: loadVoiceSkillMd(),
+      dir: "sellable-create-post",
+      displayName: "Sellable Create Post",
+      description: "Capture ideas and draft LinkedIn posts in your voice",
+      skillMd: createPostSkillMd(),
     },
   ];
 }
@@ -1700,7 +1734,22 @@ function claudeCustomAgents() {
 }
 
 function writeCodexPluginSkills(pluginRoot, opts) {
-  for (const skill of codexPluginSkills()) {
+  const skills = codexPluginSkills();
+  const currentSkillDirs = new Set(skills.map((skill) => skill.dir));
+  const skillsRoot = join(pluginRoot, "skills");
+
+  if (existsSync(skillsRoot)) {
+    for (const entry of readdirSync(skillsRoot, { withFileTypes: true })) {
+      if (!entry.isDirectory() || currentSkillDirs.has(entry.name)) continue;
+      const stalePath = join(skillsRoot, entry.name);
+      logVerbose(
+        `${C.grey}Removing stale Codex plugin skill ${stalePath}${C.reset}`
+      );
+      if (!opts.dryRun) rmSync(stalePath, { recursive: true, force: true });
+    }
+  }
+
+  for (const skill of skills) {
     const skillRoot = join(pluginRoot, "skills", skill.dir);
     writeFile(join(skillRoot, "SKILL.md"), skill.skillMd, opts);
     if (skill.soulMd) {
@@ -2500,6 +2549,8 @@ function printNextSteps(installedHosts, authReused) {
   if (hasClaude) {
     printAgentBox("Using Claude Code?", "claude", [
       { label: "Campaign", command: "/sellable:create-campaign" },
+      { label: "Post", command: "/sellable:create-post" },
+      { label: "Interview", command: "/sellable:interview" },
     ]);
     console.log("");
     console.log("");
@@ -2507,6 +2558,8 @@ function printNextSteps(installedHosts, authReused) {
   if (hasCodex) {
     printAgentBox("Using Codex?", "codex", [
       { label: "Campaign", command: "$sellable:create-campaign" },
+      { label: "Post", command: "$sellable:create-post" },
+      { label: "Interview", command: "$sellable:interview" },
     ]);
     console.log("");
   }
@@ -2777,7 +2830,11 @@ async function main() {
       console.log(`  apiUrl: ${apiUrl}`);
       console.log(`  Continue in your agent:`);
       console.log(`    Claude Code: /sellable:create-campaign`);
+      console.log(`    Claude Code: /sellable:create-post`);
+      console.log(`    Claude Code: /sellable:interview`);
       console.log(`    Codex: $sellable:create-campaign`);
+      console.log(`    Codex: $sellable:create-post`);
+      console.log(`    Codex: $sellable:interview`);
       process.exit(0);
     }
     if (rawArgs[0] === "uninstall") {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@sellable/install",
-  "version": "0.1.179",
+  "version": "0.1.181",
   "type": "module",
   "description": "One-command installer for Sellable MCP in Claude Code and Codex",
   "bin": {

package/skill-templates/create-campaign.md

@@ -408,16 +408,18 @@ automation just because a watch link exists. If `create_campaign` returns
 `watchHandoff.markdown`, print that exact value once, directly before the brief
 approval question. It will use the URL mode to say Codex or Claude Code:
 
+````markdown
 ```text
-╔══════════════════════════════════════════════════════════════╗
-║    OPEN THIS LINK TO WATCH CODEX BUILD THE CAMPAIGN LIVE    ║
-╚══════════════════════════════════════════════════════════════╝
++------------------------------------------------------+
+|         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 Markdown link is intentional: rendered chat clients turn it into a large
 click target, and plain terminals still expose the tokenized URL inside the