Install
$ agentstack add skill-viktorbezdek-skillstack-structured-writing ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Structured Writing
Most writing fails by burying the point. A structured document leads with the conclusion, supports it with grouped evidence, and lets the reader stop reading the moment they have what they need.
Pick a structure
| Structure | Best for | Lead | |---|---|---| | BLUF (Bottom Line Up Front) | Emails, status updates, requests | The ask or the answer | | Minto Pyramid | Reports, strategy memos, analyses | The single governing thought | | Inverted pyramid | News-style updates, announcements | The headline fact | | SPQR (Situation-Problem-Question-Resolution) | Briefs, consulting narratives | The situation, built to the ask |
Pick by reader, not by habit. A busy executive needs BLUF; a peer reviewing analysis needs a pyramid; a team reading an announcement needs inverted pyramid.
BLUF — the default for async work
A BLUF message opens with: the ask, the answer, or the conclusion — in under 30 words — followed by supporting detail the reader can skip.
BLUF: [ask / decision / answer in one sentence]
Why it matters:
- [reason 1 in ≤12 words]
- [reason 2]
Details (only if the BLUF doesn't answer their question):
- [supporting fact with number]
- [supporting fact with reference]
What I need from you:
- [explicit ask with deadline] or "FYI — no action needed."
Rules:
- The subject line is the BLUF, compressed.
- The BLUF sentence is self-contained — it must make sense without the rest.
- If someone replies from the first 30 words, the message worked.
Full worked examples and email/Slack/standup templates in references/bluf-templates.md.
The Minto Pyramid
Barbara Minto's pyramid: every document has one governing thought, supported by 3-5 grouped arguments, each supported by evidence.
Governing thought
(what + why + so-what)
|
┌───────────────┼───────────────┐
Argument 1 Argument 2 Argument 3
| | |
Evidence Evidence Evidence
Rules:
- Arguments at the same level share a logical relationship (MECE — mutually exclusive, collectively exhaustive).
- Each argument answers a "why" or a "how" that the governing thought raises.
- If an argument doesn't raise a predictable question, it doesn't belong at that level.
See references/minto-pyramid-method.md for the question-answer logic, MECE groupings, and the SCQ (Situation-Complication-Question) introduction.
SPQR for briefs
When the reader needs context before the ask:
- Situation — one paragraph of shared, uncontroversial context.
- Problem — the change or complication that makes a decision necessary.
- Question — the specific question you need answered.
- Resolution — your recommendation + what you need from the reader.
Useful for consulting-style briefs, board materials, vendor proposals. Less useful for async team updates (use BLUF instead).
Skim-readability checklist
Structured writing is skim-tested, not read-tested. Before sending, ask:
- [ ] Can the reader get the main point in the first 30 words?
- [ ] Does every paragraph have a clear topic sentence?
- [ ] Are supporting lists parallel in structure?
- [ ] Can the reader stop reading at any paragraph and still have the main point?
- [ ] Does the ask (if any) appear in the first sentence AND the last sentence?
- [ ] Is the longest paragraph under 100 words?
If any answer is no, restructure before polishing words.
Anti-patterns
- Buried lede — the conclusion is in paragraph 4. Move it to paragraph 1.
- Chronological narrative — "here's what we did, then what we discovered, then what we decided." Invert: lead with the decision.
- Argument without governing thought — three arguments pointing in different directions with no synthesizing claim.
- Evidence dumping — numbers and screenshots with no argument they support.
- Parallel violations — some list items are nouns, others are verbs, others are sentences. Make them parallel.
- Structure for its own sake — applying BLUF to a 200-word email nobody would read twice anyway. Structure serves speed, not ceremony.
Workflow
- Identify the reader. Who reads this, how fast, with what in their head?
- Pick the structure. BLUF for action, pyramid for analysis, inverted for announcement, SPQR for briefs.
- Write the single governing thought. If you can't state it in one sentence, you're not ready to write.
- List supporting points. 3-5, parallel, MECE.
- Fill in evidence. Numbers, references, quotes.
- Polish after structure is right. Clarity editing comes last — see
clarity-editing.
References
| File | Contents | |---|---| | references/bluf-templates.md | BLUF templates for email, Slack, standup updates, PR descriptions with worked examples | | references/minto-pyramid-method.md | Full Minto method — governing thought, MECE arguments, SCQ intros, question-answer logic | | references/structure-selection-matrix.md | Reader × purpose × medium matrix for picking the right structure |
Related skills
- clarity-editing — line-level editing once the structure is right.
- stakeholder-alignment — when the piece is an RFC/proposal, structure plus roles.
- documentation-discipline — when to write at all, and which artifact type to use.
- visual-communication — when a diagram replaces paragraphs.
- storytelling-for-stakeholders — narrative-heavy pitches and board materials.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: viktorbezdek
- Source: viktorbezdek/skillstack
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.