Skip to content

AgriciDaniel/claude-repurpose / skills/repurpose-telegram

repurpose-telegram

Verified

Generates Telegram Channel content from content atoms: structured channel posts, formatted deep dives, and quiz-ready polls. Optimizes for markdown-rich formatting, information density, and forward-ability. Sub-skill of the Content Repurposing Engine. Use when user says "telegram", "telegram channel", "telegram post", "telegram broadcast", or "repurpose for telegram".

Telegram Channel Content Generator

Generate a channel post, formatted deep dive, and poll from content atoms.

Inputs

You receive from the orchestrator (repurpose-broadcast):

InputDescription
atomsFull list of content atoms with types and impact ratings
main_argumentOne-sentence thesis of the source content
target_audienceWho benefits from this content
primary_topicCategory or niche
voice_profileDetected or overridden brand voice
brief_modeIf true, produce channel post only

References

Load before generating:

  • references/platform-specs.md -- character limits, formatting, algorithm signals
  • references/hook-formulas.md -- headline and editorial hook formulas
  • references/voice-adaptation.md -- Telegram tone rules (structured, editorial, dense)

Platform Rules

SpecValue
Channel post limit4096 chars
FormattingFull markdown (bold, italic, code, links) + HTML subset
ImagesUnlimited per post, any ratio
PollsUp to 10 options, 100 chars each, quiz mode, anonymous/public
ButtonsInline URL buttons (up to 8 per row)
Silent messagesSend without notification (supported)
Scheduled postsNative scheduling
Forward visibilityPosts forwarded to other channels/groups
Best time9-11 AM and 7-9 PM

Core Principle: Curated Feed

Telegram channels are curated feeds. Subscribers expect consistent, well-formatted, information-dense posts. Quality of formatting matters -- Telegram users notice and appreciate well-structured markdown. Forward-ability is a growth lever: design every post to be worth sharing.

Output 1: Channel Post

File: telegram/post.md

Length: 500-1000 chars.

Structure:

  1. Bold headline -- Editorial headline using **bold**. Must work as standalone hook.
  2. Key insight paragraphs -- 2-3 short paragraphs from highest-impact atoms. Use bold key terms, italic for emphasis, code for tools/commands.
  3. Takeaway list -- 3-5 actionable points. Bold the lead phrase of each item.
  4. Link button suggestion -- Inline URL button for CTA. Button text describes outcome, not action.

Rules:

  • Bold key terms aggressively -- Telegram readers scan
  • One idea per paragraph, max 3 sentences
  • Every post must deliver standalone value
  • Link buttons preferred over inline links for CTAs

Output 2: Formatted Deep Dive

File: telegram/deep-dive.md

Length: 1000-2000 chars. Only generate for high-value content with 5+ atoms.

Structure:

  1. Headline -- Bold, editorial, slightly longer than channel post headline.
  2. Problem/context -- 2-3 sentences. Use italic for the key question.
  3. Insight with evidence -- 2-3 paragraphs from stat, insight, and contrarian atoms. Bold data points and key conclusions.
  4. Actionable steps -- Numbered list (1-5 steps). Bold lead phrase + one sentence explanation per step. Must be concrete and immediate.
  5. CTA with inline button -- Suggested URL button ("get the full breakdown").

Rules:

  • Justify the length with density -- every paragraph adds new information
  • Use all available formatting (bold, italic, code, lists)
  • Numbered steps mandatory for actionable content
  • Should feel like a mini-article, not a stretched post

Output 3: Poll

File: telegram/poll.md

SpecConstraint
Options3-5 (under 100 chars each)
Max options10 (platform limit)
Quiz modeOptional -- mark correct answer with explanation
VotingAnonymous for opinions, public for quizzes

Poll structure:

  1. Setup message -- 1-2 sentences of context posted before the poll.
  2. Question -- Specific and debatable. From the content's central tension or common misconception.
  3. Options -- Defensible positions. Include one that challenges conventional thinking. Order from most to least expected.
  4. Quiz mode (optional) -- For content with a clear correct answer. Include 1-sentence explanation for the correct option.

Tone Rules

DoDo NOT
Use editorial, structured voiceWrite casually like a text message
Bold key terms and data pointsLeave text unformatted
Use numbered lists for stepsUse bullets for sequential actions
Format aggressively (bold, italic, code)Rely on plain text
Write information-dense paragraphsPad with filler or pleasantries
Design posts worth forwardingWrite self-referential content
Use inline button suggestions for CTAsBury links in paragraph text

Output Format

# Telegram Channel Post: [Topic]

**[Bold headline]**

[Formatted post body with markdown]

---
**Posting notes:** [best time, silent message suggestion, button configuration]

Brief Mode

If --brief is active, produce only:

  • 1 channel post (highest-impact atoms, standard format)
  • Skip deep dive and poll

Error Handling

ConditionAction
Fewer than 3 atomsShorter post (300-500 chars); skip deep dive
No stat or insight atomsUse howto + quote atoms; frame as practical guide
brief_mode is trueGenerate channel post only; skip deep dive and poll
Content too casualIncrease structure; add bold formatting and lists
Content is promotionalReframe as editorial insight; strip product mentions
Long content (10+ atoms)Split into channel post + deep dive
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 467kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 319kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 319kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 319kUpdated Claude CodeCodexVerifiedInstall