Skip to content
OpenAgentSkill
Browsing is open · Catalog captured 2026-11-21 · Automatic updates and account services are paused.
Design & UI

content-repurposer

Turn one long artifact into distinct voice-matched posts. Use for a week of content from a talk, article, or call. Read README.md for the guide map.

Mario Aldayuz / Littlebird · Static checks on file

Installation

Read the source and dependencies before running a command. Copying does not install or execute anything.

npx skills add legioncodeinc/vibe-coding-tools --skill content-repurposer

About this skill


name: content-repurposer description: "Turn one long artifact into distinct voice-matched posts. Use for a week of content from a talk, article, or call. Read README.md for the guide map." license: AGPL-3.0-or-later compatibility: Claude Cowork, Claude Code 2.1 or newer, Cursor 2.4 or newer, Codex metadata: hive-tier: standalone version: "1.0.0" author: "Mario Aldayuz / Littlebird" requires: "Littlebird MCP (Power or Pro plan)"

Content Repurposer

Purpose

One good long-form artifact becomes a week of derivatives. A Facebook post, a call transcript, a webinar, a Loom, a long email. The output is a content pack: 5 to 7 short posts that each stand alone, a carousel or slide outline with per-slide copy, an email, a thread, optionally a short-video script or hook list, a sequencing plan that says which piece goes out when and why, and a do-not-publish list.

The craft problem this skill exists to solve. Most repurposing output is the same idea restated five times in different shapes. To anyone who follows the user on more than one surface, that reads as spam, and that reader is the one the user most wants to keep.

The fix is not a better hook per format. Each derivative takes a DIFFERENT ANGLE on the source, not a different format of the same angle. The one controlled experiment in the research archive distinguishes cosmetic variation, meaning the same message in altered packaging, from substantive variation, meaning a change in the arguments themselves, and finds that cosmetic variation carried only under LOW message relevance while substantive variation carried with an audience motivated to process it (references/research/distilled-repurposing-and-format-adaptation.md, section 1). A multi-surface follower is the motivated case by construction.

references/angle-taxonomy.md is the guide that makes this operational, and it is the most important file in the skill.

It drafts. It never posts, and it never publishes.

Relationship to said-it-already

These are siblings, not duplicates. said-it-already MINES many sources for seeds. content-repurposer takes ONE chosen artifact and expands it into a week. Where their craft overlaps, this skill points rather than restates: the spoken-to-written rebuild, attribution screening, and the confidentiality screen all live in that skill's guides and are used from there.


Littlebird MCP calls used

Real tool names, verified against the live server (references/littlebird-mcp-reference.md). LIST the tools available in this session before calling anything and use the names you actually find.

ToolUsed for
search_user_contextFinding a described post or written artifact the user cannot link, and sweeping for how the original performed
LB_INTERNAL_SEARCH_MEETINGSLocating a call by TOPIC when the source is a conversation
LB_INTERNAL_LIST_MEETINGSLocating a call by NAME or date, and enumerating a window
LB_INTERNAL_GET_MEETINGThe structured summary, which carries owner attribution
LB_INTERNAL_GET_MEETING_TRANSCRIPTExact wording only, never attribution
LB_INTERNAL_GET_ROUTINE_REPORTSReading the weekly nomination routine's own past reports before selecting a source
LB_INTERNAL_LIST_ROUTINESChecking whether the nomination routine already exists
LB_INTERNAL_GET_ROUTINE_CONFIGReading the existing routine before any update
LB_INTERNAL_CREATE_ROUTINECreating the weekly nomination routine, from an interactive session only
LB_INTERNAL_UPDATE_ROUTINEChanging it later, from an interactive session only
LB_INTERNAL_GET_SUBSCRIPTION_STATUSExplaining a plan gate or a routine-count limit

There is no Littlebird tool that searches past Littlebird chat conversations. Where a source lives in one, use search_user_context and say so.


Trigger

On demand is the primary mode. Trigger when the user says any of: repurpose this, turn this into a week of content, make content out of this call, break this down for LinkedIn, I need a content pack, what else can I do with this webinar, get more out of this post.

The optional weekly routine is a nomination mode. It does not produce the pack. It names a candidate artifact and explains why, then hands off.


Capability gate

This skill requires the Littlebird MCP on a Power or Pro plan.

  1. LIST the tools actually available in this session. Do not assume tool names.
  2. If no Littlebird tools are present, stop and tell the user the skill needs the Littlebird MCP connected, with a link to https://support.littlebird.ai/docs/mcp/.
  3. If the tools are present but return plan errors, call the subscription status tool and report the plan gate.

One exception worth naming: if the user pastes the source artifact directly, the skill can run without retrieval. Say so, and say what is lost, which is the performance sweep and any surrounding context.


Routine cadence

On demand is primary. The weekly routine is optional and it only nominates.

Schedule: weekly, {"frequency": "weekly", "time": "09:00", "week_days": ["MO"]} in the user's local timezone, so the nomination lands at the start of a working week with time to build the pack.

The routine cannot produce the pack. Building a pack requires a confidentiality screen, an angle audit, and an approval gate on the actual final text, and a routine cannot run an approval gate (references/littlebird-mcp-reference.md, the Routines-observe Cowork-acts pattern). Exact prompt text is in the routine wiring section below.


Process

Step 1: Frame the run

Ask with AskUserQuestion, in one batch:

  1. The source. A link, a paste, a description to search for, or a call to find. If they do not have one and the nomination routine exists, read its latest report first.
  2. The surfaces. Which platforms they publish to, plus email yes or no. This sets the format constraints and t
Read the complete source ↗