Agent skill

Storytelling Narrative

by sammcj in sammcj/agentic-coding

Structure, refine, or draft narrative content or storytelling for business communication such as slide decks, blog posts, pitches, reports.

Apache-2.0Auto-check passedWriting & Content

Install Storytelling Narrative

skills CLI
$ npx skills add sammcj/agentic-coding --skill storytelling-narrative -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install sammcj/agentic-coding storytelling-narrative --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/sammcj/agentic-coding.git skills-src && mkdir -p .claude/skills && cp -r skills-src/Skills_disabled/storytelling-narrative .claude/skills/storytelling-narrative && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
storytelling-narrative
GitHub stars
162
Token cost
~3.6k tokens
SKILL.md length
2,027 words
Files
5 (incl. references)
Skills in repo
64
Repo updated
First seen
Licence
Apache-2.0

At a glance

Structure, refine, or draft narrative content or storytelling for business communication such as slide decks, blog posts, pitches, reports.

  • Works in 8 steps: Get oriented → Engage with the material → Find the spine → …
  • The user wants help building a narrative arc
  • SKILL.md covers Operating principles, Narrative primitives, Workflow and When to use which reference, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Storytelling Narrative is an agent skill from sammcj/agentic-coding. Structure, refine, or draft narrative content or storytelling for business communication such as slide decks, blog posts, pitches, reports. Use when the user wants help building a narrative arc, find the through-line, sequence beats or refine existing content for flow and impact.

Its SKILL.md is about 3.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including reference files (for example `references/critique.md`, `references/formats.md` and `references/hooks-and-moments.md`).

It sits in Writing & Content, covering Storytelling, Blog and article writing and Slides and decks. The repository describes itself as: Agentic Coding Rules, Templates etc... The licence is Apache-2.0.

When your agent uses it

  • The user wants help building a narrative arc
  • Find the through-line
  • Refine existing content for flow and impact

Example prompts

  • “/storytelling-narrative”

Workflow steps

8 steps, taken from the step headings in SKILL.md.

  1. Get oriented
  2. Engage with the material
  3. Find the spine
  4. Choose a structure (lens)
  5. Order the beats
  6. Decide output mode
  7. Draft (if requested)
  8. Self-review

What it can do on your machine

Read from SKILL.md and the folder at commit 2f25ced. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Storytelling Narrative loads about 3.6k tokens when it runs, and up to ~14k if it reads all its reference files. Until then it costs about 76 tokens; SKILL.md has 2,027 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~76
When it runs · the whole SKILL.md, loaded when a task matches
~3.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~14k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from sammcj/agentic-coding at commit 2f25ced, republished under its Apache-2.0 licence (© sammcj). 2,027 words, ~3,634 tokens.

Download SKILL.mdSave it as .claude/skills/storytelling-narrative/SKILL.md (or your agent's skills folder). This skill also uses 4 other files; get the full folder from GitHub.
name
storytelling-narrative
description
Structure, refine, or draft narrative content or storytelling for business communication such as slide decks, blog posts, pitches, reports. Use when the user wants help building a narrative arc, find the through-line, sequence beats or refine existing content for flow and impact.

Storytelling & Narrative

Help the user build, refine, or restructure a narrative for a business artefact: slide deck, blog post, pitch, exec memo, article, or report. The skill is format-agnostic at its core. Format-specific notes live in references/formats.md.

The user is a domain expert. They usually know the material deeply or have rough touchpoints in mind. What they're missing is structure, flow, and a clear through-line. Your job is to help them surface what's already there and shape it, not to invent narrative from nothing or layer marketing language on top.

Operating principles

Lenses, not templates. Storytelling frameworks (Pixar, ABT, Three-Act, StoryBrand, Hero's Journey, Golden Circle, SCQA, Pyramid Principle, Dykes Arc, Duarte Sparkline) are different surface arrangements of the same underlying primitives: status quo, tension, stakes, turn, resolution. Treat named frameworks as lenses to check a story against, not templates to fill in. Pick a lens when it fits the audience-goal-material combination - never force material into a framework that doesn't match.

Narrative comes from the material, not a script. Most business storytelling failures aren't structural - they're the result of not engaging with the material long enough to find what's actually there. Spend real time in Phase 2 (engaging the material) before reaching for any framework.

Anti-slop posture. No marketing adjectives, no hot takes, no cliche verbiage or buzzwords. Concrete nouns over abstract ones. Specific people, specific moments, specific numbers. If a sentence could appear in any company's blog post, rewrite it.

Pause when there's a real fork; proceed when there isn't. Use multi-choice questions (the AskUserQuestion tool) when the user genuinely needs to steer - multiple plausible spines, multiple framework lenses, multiple orderings, output-mode choice. Don't pause on small calls or things you can infer with confidence. The user has explicitly said: increase autonomy when you have understanding, ask when you don't.

Narrative primitives

Every narrative - regardless of framework - is built from these. Find them in the material before structuring anything.

  • Status quo / "what is" - the world the audience already lives in. Familiar enough that they nod.
  • Tension - what's wrong, missing, or about to break. The reason the story exists.
  • Stakes - what happens if nothing changes. Without stakes, the audience has no reason to care.
  • Turn - the moment, decision, evidence, or insight that changes things. The most often-missing primitive in weak business narratives.
  • Resolution - the new state, the action, the call. What the audience now does, believes, or decides.
  • Hook - the entry point that opens the loop. Must land within the first 30-60 seconds (or first paragraph). Six common types in references/hooks-and-moments.md.
  • Through-line - a single sentence the rest of the piece serves: "from X to Y because of Z". If you can't write the through-line, the narrative isn't ready.

The shape is always: old understanding → tension → new understanding → new action. Different frameworks just arrange this differently.

Workflow

This is a phased flow, not a rigid sequence. Skip phases the user has already done; double back when new information surfaces. The pause points below are where multi-choice user input genuinely steers the work.

Phase 1: Get oriented

Don't start structuring until you understand audience, goal, material state, and format. This isn't a "quick diagnostic" - these answers shape everything downstream. Ask substantively, but with focus.

Things you need to know:

  • Audience. Who is this for? What do they already know? What do they care about? What position or fatigue are they bringing in?
  • Goal. What should the audience think, feel, or do after consuming this? A specific decision, a belief shift, an action, an emotional response?
  • Material state. Raw knowledge in their head? Rough touchpoints sketched out? An existing draft to refine? An existing deck/data to reshape?
  • Format and constraints. Slide deck, blog post, pitch, memo, article, report? Length? Time-to-deliver? Live presentation or read-alone?
  • Tone. Visionary / pragmatic / friendly / urgent / credible / sceptical / confident.

Use AskUserQuestion for things with discrete plausible answers (tone, audience type, format, material state). Use open prose questions when you need substance (audience-specific concerns, goal nuance).

If the user volunteered most of this in their initial prompt, don't re-ask - confirm what you understood and ask only what's missing.

Phase 2: Engage with the material

This is the most undervalued phase. Skip it and you'll produce a structure that doesn't fit what the user actually has to say.

Branch by material state:

Raw knowledge / rough touchpoints - ask the user to talk through what they have. Don't summarise back yet - keep extracting. Probe for narrative substrate:

  • "What changed? What made you start paying attention to this?"
  • "What was the moment you realised X?"
  • "Who's affected? What do they actually do differently?"
  • "What's the part most people get wrong?"
  • "What number or specific example surprised you?"

You're looking for moments, not abstractions. Concrete things - a meeting, a number, a customer, a failure, a near-miss - are the raw material narratives are built from.

Existing draft - read carefully. Find the implicit arc (or note its absence). Mark:

  • Where the hook is (or where it should be)
  • Whether there's a turn - and where
  • What's the spine the rest serves
  • What's repeated, what's buried, what's missing

Read references/critique.md for the diagnostic checklist.

Existing data / deck - look for the transformation already present. The story is usually there waiting to be uncovered: a before-after, a problem-solution, a counterintuitive finding, a shift in trend. Resist the urge to invent a story; surface the one that's there.

Output of this phase: a list of raw narrative material - concrete moments, facts, characters, tensions, changes - not yet ordered.

Phase 3: Find the spine

Distill the raw material to:

  • A single-sentence through-line: "from X to Y because of Z"
  • A candidate hook
  • The turn - what shifts
  • The stakes - why it matters if nothing changes
  • The resolution - new state, new action

If you can find more than one plausible spine, that's a fork - present 2-3 options to the user via AskUserQuestion and let them pick. Different spines lead to genuinely different narratives; this is one of the highest-impact decision points.

If there's only one spine that fits the material, propose it directly with rationale. Don't manufacture options for the sake of asking.

Phase 4: Choose a structure (lens)

Match a framework lens to the audience-goal-material combination. See references/narrative-frameworks.md for the full catalogue and selection guidance. Quick orientation:

  • ABT (And/But/Therefore) - short pieces: exec updates, elevator pitches, paragraph openers, slide subtitles.
  • Three-Act - general default for blog posts, talks, longer writing.
  • Pixar (Once upon a time...) - change stories, adoption stories, "how X became Y".
  • Golden Circle (Why/How/What) - vision, purpose, mission, manifesto.
  • StoryBrand SB7 - customer-facing marketing where the customer is the hero.
  • Hero's Journey (condensed) - founder origin, brand history, personal narrative.
  • Duarte Sparkline (what is / what could be) - persuasive keynotes, change comms.
  • Dykes Data Storytelling Arc - data-driven presentations and insight reports.
  • SCQA (Situation/Complication/Question/Answer) - exec briefings, consulting memos.
  • Pyramid Principle (Minto) - top-down written reports for executives who scan.

Lenses are not exclusive - a piece can use SCQA at the document level and ABT at the paragraph level. Present 2-3 lens options when more than one would work; explain the trade-offs briefly; let the user pick.

Show full SKILL.md (864 more words)Show less
Phase 5: Order the beats

Arrange the material into the chosen structure. Three things to attend to:

  • Sequence for emotional and logical flow. The audience should always know why they're reading the next paragraph. Each beat earns the next.
  • Cut ruthlessly. Most business narratives carry 30-50% material that doesn't serve the through-line. If a beat doesn't move the audience from where they were to where they need to be, it goes. Mark cuts as "park for later" rather than deleting outright - the user may reuse them elsewhere.
  • Place the turn. The turn is usually mid-to-late, not upfront. Resist "executive summary" instinct that puts the conclusion first - that kills the narrative tension. Exception: read-alone formats for time-poor execs (memos, reports) often do need bottom-line-up-front; in that case, give a one-sentence headline then tell the story.

Show the user the proposed ordering. Pause via AskUserQuestion only if there are multiple genuinely plausible orderings - otherwise propose and let them react.

Phase 6: Decide output mode

Before drafting anything, ask the user what level of help they want. Use AskUserQuestion with options like:

  • Structure only - they take it from here and write the prose themselves.
  • Hook + key moments - draft the opener and the most load-bearing beats; they fill the rest.
  • Full draft - draft prose for every beat in their voice.
  • Refine existing draft to new structure - rework what they already wrote against the new spine.

This is non-negotiable as a pause point. Don't assume the level of involvement the user wants.

Phase 7: Draft (if requested)

When drafting, hold to:

  • Concrete over abstract. Specific nouns, specific numbers, specific people. "The auth team rewrote the session middleware in two weeks" beats "The team transformed our authentication".
  • Verbs do the work. Strong verbs over verb+adjective combos.
  • Vary sentence length. Mix short with long. Don't write five sentences of the same length and structure in a row.
  • The user's voice, not yours. Match the tone they signalled in Phase 1 and the rhythm of any existing material they shared.
  • Anti-slop check. No marketing adjectives, no AI tells, no "Additionally" / "Furthermore" / "It's worth noting" sentence openers, no closing summaries that say "In conclusion, this approach...".
  • Show stakes and consequences. Don't gloss over the tension - that's where the narrative energy lives.

For format-specific drafting concerns (slide one-idea-per-slide rule, blog opener strength, pitch tightness), read references/formats.md.

Phase 8: Self-review

Before returning the final output, run a critique pass against references/critique.md. Common failure patterns:

  • Hook is generic or buried
  • Turn is missing or under-emphasised
  • Stakes implied but not stated
  • Through-line drifts mid-piece
  • Slop language crept in during drafting

Fix these before delivering. If you find structural issues, surface them rather than silently rewriting - the user may have specific intent behind a choice that looks weak.

When to use which reference

  • references/narrative-frameworks.md - when picking a lens, or when the user wants to commit to a specific framework. Selection guidance, per-framework failure modes, and detail on the lesser-known lenses.
  • references/hooks-and-moments.md - when crafting a hook or sharpening a turn. Six hook types, the turn test, placement principles.
  • references/formats.md - when format-specific concerns matter (slide deck rules, blog opener, pitch arc length, memo bottom-line-up-front).
  • references/critique.md - when reviewing existing content, or running self-review on drafted output.

Multi-choice question patterns

The user has explicitly asked you to use AskUserQuestion where it earns its place. Heuristics for when to pause:

Pause forDon't pause for
Tone (visionary / pragmatic / urgent / credible) when not signalledWord-level choices
Audience type when ambiguousTiny edits
Multiple plausible spines (genuine fork)Obvious cuts
Framework lens with 2-3 viable fitsSingle-fit lens
Ordering when multiple orderings are validMechanical sequencing
Hook variant when 2-3 are equally plausibleWhen one hook clearly fits

Always pause for output mode (structure-only / hook + key moments / full draft / refine existing draft) before drafting. Don't infer what level of help the user wants.

Ask 1-3 questions at a time, not 5+. Let the user steer; don't quiz them.

Gotchas

  • Don't fall back to the same handful of frameworks every time. Pixar, Three-Act, Golden Circle, StoryBrand, Hero's Journey, ABT are widely known, but they're a subset. SCQA, Pyramid Principle, Dykes Data Storytelling Arc, and Duarte's Sparkline are often the better fit for exec, consulting, data-driven, or report-style work. Pick by audience-goal-material fit, not by familiarity.
  • This skill is narrative-first and format-agnostic. Charts, dashboards, and visual design are not in scope. If the piece is primarily data-driven visuals, hand off to the storytelling-with-data skill; otherwise stay on the narrative.
  • Don't skip Phase 2. Engaging with the material is what makes the rest work. Frameworks applied to thin material produce slop.
  • Don't manufacture options to ask the user. If there's one obvious answer, propose it. Multi-choice questions are for genuine forks, not for performance of consultation.
  • Don't lecture the user about narrative theory. They want help with their content, not a lesson. Apply the lens silently; surface only what they need to make decisions.
  • Don't add closing "In summary..." paragraphs to drafted content. That's a slop tell. The narrative should land naturally on its resolution.
  • The turn is the most under-built primitive in business writing. Most weak drafts have a status quo, stakes, and a vague resolution but no clear shift. Watch for this when reviewing existing content.

© sammcj, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 4 other files (references) in Skills_disabled/storytelling-narrative of sammcj/agentic-coding.

  • SKILL.md
  • references/critique.md
  • references/formats.md
  • references/hooks-and-moments.md
  • references/narrative-frameworks.md

Open the folder on GitHubat commit 2f25ced

Compare with similar skills

Storytelling Narrative next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Storytelling Narrative compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Storytelling Narrative this skillsammcj/agentic-coding162—~3.6kAutomated safety check: PassApache-2.0
Data Storytellingynulihao/AgentSkillOS61815 repos~2.8kAutomated safety check: PassNone
Reportmicrosoft/data-formulator18k—~1.5kAutomated safety check: PassMIT
Blog Cover Posterdigoal/blog8.6k—~1.4kAutomated safety check: PassGPL-2.0
Editorial Illustrationshuytieu/COG-second-brain1.3k—~2.4kAutomated safety check: PassMIT
Story Bank Builderaaron-he-zhu/aaron-marketing-skills2.9k—~3.2kAutomated safety check: PassApache-2.0

Similar skills

  • Data Storytelling

    ynulihao/AgentSkillOS

    Transform data into compelling narratives using visualization, context, and persuasive structure.

    618 GitHub starsUsed in 15 repos~2.8k tokens
    Writing & ContentAuto-check passed
  • Report

    microsoft/data-formulator

    Official

    Turn an exploration (threads, findings, charts) into a single Markdown report — note, blog post, executive summary, KPI dashboard, slide brief, or multi-section analytical report, with embedded…

    18k GitHub stars~1.5k tokensUpdated 3 days ago
    Writing & ContentAuto-check passed
  • Turn a blog post, article, notes, or any source material into a set of vertical poster images — one cover plus several coherent content slides that explain the core points.

    8.6k GitHub stars~1.4k tokensUpdated 2 days ago
    Writing & ContentAuto-check passed
  • Editorial Illustrations

    huytieu/COG-second-brain

    Generate meaning-carrying editorial data-illustrations in the monotykamary / Linear aesthetic (near-black grayscale, Inter display + mono labels, hairline framed figures) with a single coral accent.

    1.3k GitHub stars~2.4k tokensUpdated 8 days ago
    Writing & ContentAuto-check passed
  • Story Bank Builder

    aaron-he-zhu/aaron-marketing-skills

    A skill your agent uses when the user asks to "build a story bank", "collect our origin and customer stories", or "assemble reusable proof stories for the message"; assembles reusable narrative…

    2.9k GitHub stars~3.2k tokensUpdated yesterday
    Writing & ContentAuto-check passed
  • Linkedin Post Generator

    Varnan-Tech/opendirectory

    Converts any content, blog post URL, pasted article, GitHub PR description, or a description of something built, into a formatted LinkedIn post with proper hook, story arc, and formatting.

    674 GitHub stars~2.6k tokensUpdated 1 mo ago
    Writing & ContentAuto-check: notes

More from sammcj/agentic-coding

All 64 skills in this repo
  • Yue2 Music

    sammcj/agentic-coding

    A skill your agent uses when generating songs with YuE2, covering a recording via SheetSage2 audio-to-ABC, editing a score or lyrics with melody preservation, or building a reproducible listening…

    162 GitHub stars~2.3k tokensUpdated 2 days ago
    Auto-check passed
  • Bento Slides

    sammcj/agentic-coding

    A skill your agent uses when creating or editing Bento (.bento.html) slide decks, including any request for a single-file HTML slide deck.

    162 GitHub stars~2.9k tokensUpdated 2 days ago
    Auto-check passed
  • Idrive Backup

    sammcj/agentic-coding

    A skill your agent uses whenever the user wants you to manage, discuss or diagnose iDrive Backup configuration on macOS

    162 GitHub stars~1.7k tokensUpdated 2 days ago
    Auto-check: notes
  • Piper Tts Training

    sammcj/agentic-coding

    Train custom TTS voices for Piper (ONNX format) using fine-tuning or from-scratch approaches.

    162 GitHub stars~1.4k tokensUpdated 2 days ago
    Auto-check passed
  • PPTX To Md

    sammcj/agentic-coding

    Convert a PPTX slide deck into per-slide markdown that preserves both the verbatim text and the meaning of embedded screenshots, diagrams and charts in their original layout positions.

    162 GitHub stars~1.8k tokensUpdated 2 days ago
    Auto-check passed
  • Skill Creator Primer

    sammcj/agentic-coding

    You MUST load this skill before the skill-creator skill AND before making ANY change to, or conducting a review of ANY Agent Skill.

    162 GitHub stars~9.8k tokensUpdated 2 days ago
    Auto-check passed

Questions about Storytelling Narrative

What does Storytelling Narrative do?

Structure, refine, or draft narrative content or storytelling for business communication such as slide decks, blog posts, pitches, reports. Storytelling Narrative is an agent skill from sammcj/agentic-coding. Structure, refine, or draft narrative content or storytelling for business communication such as slide decks, blog posts, pitches, reports.

When should I use Storytelling Narrative?

Storytelling Narrative fits situations like: the user wants help building a narrative arc; find the through-line; refine existing content for flow and impact.

How do I install Storytelling Narrative in Claude Code?

Run `npx skills add sammcj/agentic-coding --skill storytelling-narrative -a claude-code`. Or copy the skill folder (Skills_disabled/storytelling-narrative in sammcj/agentic-coding) into .claude/skills/storytelling-narrative in your project. Claude Code loads it when a task matches its description.

How do I install Storytelling Narrative in Codex?

Run `npx skills add sammcj/agentic-coding --skill storytelling-narrative -a codex`. Or copy the skill folder (Skills_disabled/storytelling-narrative in sammcj/agentic-coding) into .agents/skills/storytelling-narrative in your project. Codex loads it when a task matches its description.

Can I use Storytelling Narrative in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add sammcj/agentic-coding --skill storytelling-narrative -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/storytelling-narrative, .gemini/skills/storytelling-narrative, .github/skills/storytelling-narrative and .opencode/skills/storytelling-narrative in your project.

What does Storytelling Narrative need to run?

SKILL.md names no scripts, command-line tools or credentials: Storytelling Narrative is instructions for the agent only.

Does Storytelling Narrative access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Storytelling Narrative safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Storytelling Narrative use?

Storytelling Narrative is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Storytelling Narrative use?

About 3.6k tokens (SKILL.md is roughly 15k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 10k tokens, read only when the agent opens those files.

What are the alternatives to Storytelling Narrative?

Skills that share tags, products or a category with Storytelling Narrative: Data Storytelling (ynulihao/AgentSkillOS, 618 stars), Report (microsoft/data-formulator, 18k stars), Blog Cover Poster (digoal/blog, 8.6k stars) and Editorial Illustrations (huytieu/COG-second-brain, 1.3k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Storytelling Narrative?

sammcj (a GitHub user) maintains it in sammcj/agentic-coding, which has 162 GitHub stars. The repository holds 64 skills in this directory. The repository was last updated on October 9, 2026.

Source: sammcj/agentic-coding on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.