agentleFS
darellchua2darellchua2/civiltekk-opencode-claude-skills/skills/horseshoe-paper-writing-skill/SKILL.md

horseshoe-paper-writing-skill

Skill7 starsChanged 59 days agoApache-2.0View on GitHub

Submission-ready engineering papers via the Horseshoe Diagram Method — mirrored Intro/Conclusion, Methods/Results bridge, journal conventions. Triggers: horseshoe paper, journal paper draft.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write submission-ready engineering papers via the Horseshoe Diagram Method: a paper is an inverted-U whose two open arms (Introduction, …

    line 11

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

What I do

Write submission-ready engineering papers via the Horseshoe Diagram Method: a paper is an inverted-U whose two open arms (Introduction, Conclusion) must mirror each other, joined by the evidence bridge (Methods → Results → Discussion).

When to use me

"horseshoe paper", "journal paper draft" — any engineering/ML journal submission draft, revision, or review under this method.

1. The method (core)

Left arm (Intro) funnels INWARD: broad context → specific problem → literature gap → contributions/RQs. Bridge is the ONLY evidence: Methods, Results, Discussion. Right arm (Conclusion) flares OUTWARD: findings → implications → limitations → future work → broader context.

The Mirror Principle (the one rule): every claim/question/contribution in the Introduction MUST have a mirrored partner in the Conclusion — no new content in the Conclusion, only answers to Intro promises. Contributions answered in the SAME ORDER; RQ1..RQn answered in order. A reviewer must be able to fold the paper along the bridge and have each intro paragraph line up with its conclusion partner. (vs "hourglass": the horseshoe makes intro/conclusion drift visible and fixable.)

2. The seven-step workflow

  1. Draft the LEFT ARM (promises, contributions, RQs) → 2. Draft the RIGHT ARM (answers, point-by-point mirror) → 3. Build the PAIRING TABLE (intro ¶ ↔ conclusion ¶) → 4. Draft the BRIDGE to connect each pair → 5. Abstract (compress the whole horseshoe, ~250 words) → 6. Front/back-matter (Practical Applications/Highlights, Data Availability, Author Contributions, Acknowledgments) → 7. MIRROR AUDIT.

Never write the bridge first — the arms define the contract; the bridge fulfills it.

3. Journal-submission section order (stable across ASCE/Elsevier/IEEE/Springer/MDPI; names vary by venue)

Abstract → Introduction → (Literature Review) → Methods → Results → Discussion → Conclusions → Practical Applications/Highlights (venue-dependent) → back-matter. Methods before Results; Discussion separate from Results unless venue merges them.

4. Mirror audit (mandatory before done)

Two-column pairing table (section-templates/pairing-table.md): Intro element ↔ Conclusion partner ↔ Bridge evidence § ↔ ✓. Rules: (1) every numbered contribution has a Conclusion partner — add one or delete the contribution; (2) no orphan RQs; (3) NO new content in Conclusion (new claim/dataset/citation = violation — move to bridge); (4) Abstract mentions every contribution; (5) Practical Applications cites the single best headline number. Run before submission, after EVERY revision, and during review-response (new analysis → bridge AND the relevant arm).

5. Evidence rules (condensed)

ML results report per-class metrics (precision/recall/F1), never accuracy-only. Implications quantify ("reduces X by 17%") or are cut. Limitations subsection always present. Generalization hedged to demonstrated scope.

6. Anti-patterns to reject (review checklist)

Methods-first writing · unmirrored contributions · orphan RQs · new content in Conclusion · missing Practical Applications/Highlights when required · accuracy-only reporting · no limitations subsection · vague implications · overstated generalization.

Related: research-paper-generation-skill (generation pipeline from codebases) · the docx route of office-document-router-skill (DOCX conversion) · civiltekk-diagram-skill (mermaid route).

Removed 2026-09: the §3 section-by-section writing guide (per-section prose templates), §4 multi-venue reference-style catalogs, §6 diagramming conventions detail, §7 formula/equation typography rules, §8 table formatting walkthroughs, §9 folder layout, §13 example invocation transcript, ASCII reference figure — venue-specific mechanics the model knows; the method, workflow, audit, and anti-patterns above are the durable contract.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.