essay_crafter
Trigger: essay, ensayo, academic essay, claim audit, citation check, reviewer.
Skill by gentleman-programming
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/fegome90-cmd-essay-crafter
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: security, research
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/essay-crafter
What's inside
- Skill name
essay_crafter- Description
- Trigger: essay, ensayo, academic essay, claim audit, citation check, reviewer. Build essays with evidence passports, ARS-style integrity gates, and OpenAI-ready structured outputs.
- License
- Apache-2.0
- Metadata
- author: gentleman-programming · version: 2.0
- Instructions
- 88 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- name: only lowercase letters, digits and single hyphens allowed
Files (19)
- SKILL.md
- evidence_passport.schema.json · assets/
- outline_template.md · assets/
- academic_register.md · references/
- conclusion_microstructure.md · references/
- counterargument.md · references/
- editorial_cleanup.md · references/
- evidence_passport.md · references/
- failure_paths.md · references/
- integrity_pipeline.md · references/
- introduction_microstructure.md · references/
- paragraph_density.md · references/
- paraphrasing.md · references/
- revision_loop.md · references/
- structure_gate.md · references/
- structured_output_complements.md · references/
- transitions.md · references/
- warrants.md · references/
- word_count.md · references/
SKILL.md (first lines)
## Activation Contract Use when writing or revising an academic essay that must survive claim, citation, and prose scrutiny. Prefer for deterministic pipelines, not freeform brainstorming. ## Hard Rules - Start by converting the topic into an arguable research question. - Define an explicit thesis and make the introduction close with a roadmap sentence. - Do not draft from loose notes. Every note MUST have `source_id` plus a locator anchor. - Keep `sources`, `notes`, `claims`, and `claim_links` separate. Claims never point directly to bibliography entries. - Keep one primary analytical dimension per section or paragraph cluster. Do not mix empirical evidence, ethics, regulation, and normative synthesis in the same block. - Each paragraph should do one rhetorical job only. Split paragraphs that change function midstream. - Reject any claim that lacks mapped evidence or only has weak contextual support. - Address at least one significant counterargument or opposing position in every essay. - Make the warrant connecting each claim to its evidence explicit. Every claim-evidence pair MUST include an explicit warrant (see `references/warrants.md`). Do not leave reasoning assumed. - Run prose calibration before final formatting: remove AI-typical phrasing, inflated certainty, and broken paragraph flow. - Block final audit if orphan tokens, truncated fragments, or leftover placeholders remain. - Paraphrase rather than quote; follow `references/paraphrasing.md`. Flag any direct quote exceeding 40 words. ## Decision Gates | If | Then | |---|---| | Topic is descriptive only | Rewrite it as a contested question before proceeding | | Note has no locator | Quarantine it; do not cite or reuse it | | Introduction lacks roadmap | Block progression to drafting | | A paragraph mixes evidence, regulation, ethics, or synthesis | Split into separate sections or paragraphs | | Claim bundles multiple assertions | Split into atomic `claim_id`s | | A paragraph performs more than one rhetorical job | Rewrite or split it | | Evidence only provides context | Use it for framing, not as primary support | | No counterargument addressed | Require one before proceeding | | Claim-evidence pair lacks explicit warrant | Flag as unsupported inference; add warrant before proceeding | | Essay type not selected | Default to argumentative | | Word count exceeds +/-10% tolerance | Flag for adjustment | | Orphan token or truncated fragment appears | Block final audit until cleaned | | Final draft changes wording materially | Re-run claim audit on affected claims | ## Execution Steps 1. Define the essay question and success criteria. Select essay type from: argumentative, comparative, chronological, thematic, problem-solution. Default: argumentative. 2. State the thesis and build the roadmap with `assets/outline_template.md`. **CP1: Thesis + Roadmap Approval** 3. Build the evidence passport using `assets/evidence_passport.schema.json`. 4. Map each note to one analytical dimension and drop tangential branches. 5. Draft long first, then run the structure gate from `references/structure_gate.md` (includes counterargument checks). 6. Run the paragraph-density pass from `references/paragraph_density.md`. 6a. Integrate counterarguments per `references/counterargument.md`. Address at least one significant opposing position using steel-man + rebuttal structure. **CP2: Structure + Counterargument Approval** 7. Simulate four reviewers: editor, method critic, devil's advocate, and architecture editor. 7a. Apply revision loop from `references/revision_loop.md`. Categorize reviewer findings as structural, evidential, or prose. Max 2 revision rounds per category. Unresolved findings become acknowledged limitations. **CP3: Revision Review** 8. Run editorial cleanup with `references/editorial_cleanup.md`. Calibrate academic register per `references/academic_register.md`. Repair broken transitions per `references/transitions.md`. 9. Compile `claims` + `claim_links` for final audit. 10. Use local audit references plus structured output complements from `references/structured_output_complements.md` when structured outputs or batch review help. ## Checkpoint Protocol Each checkpoint follows: PAUSE → PRESENT summary → WAIT for user signal. - **APPROVE**: continue to next phase - **REDIRECT**: incorporate feedback, re-run current phase - **ABORT**: stop pipeline, deliver current state Checkpoints are advisory in automated/batch mode — the pipeline continues without pausing. ## Output Contract Return: - Final research question. - Evidence passport path or object. - Outline tied to evidence clusters. - Claim-to-evidence map. - Reviewer findings and revision report. - Final audit status: pass, rework, or blocked. ## References - `assets/evidence_passport.schema.json` · `assets/outline_template.md` - `references/evidence_passport.md` · `references/structure_gate.md` - `references/paragraph_density.md` · `references/editorial_cleanup.md` - `references/integrity_pipeline.md` · `references/structured_output_complements.md` - `references/counterargument.md` · `references/revision_loop.md` - `references/failure_paths.md` · `references/warrants.md` - `references/word_count.md` · `references/academic_register.md` - `references/transitions.md` · `references/paraphrasing.md` - `references/introduction_microstructure.md` · `references/conclusion_microstructure.md`
From fegome90-cmd/paper-writer, Apache-2.0.
More from fegome90-cmd/paper-writer
- 0.463score
literature-search
fegome90-cmd
Búsqueda bibliográfica sistemática con ranking de calidad, descarga de PDFs y exportación estructurada.
SkillCFreshness: CClaude Code +9 - 0.407score
trifecta-mcp
gentleman-programming
Use for code navigation, graph impact analysis, semantic search, and context validation via Trifecta MCP.
SkillCFreshness: CClaude Code +9 - 0.247score
academic_writer
fegome90-cmd
No description from the publisher yet.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jun 5, 2026 (4 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- Version
- 2.0
- License
- Apache-2.0
- Repository
- github.com/fegome90-cmd/paper-writer
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from gentleman-programming
- 0.724score
gentleman-bubbletea
gentleman-programming
Bubbletea TUI patterns for Gentleman.Dots installer.
SkillAFreshness: A★ 7.6KClaude Code +9 - 0.723score
engram-backlog-triage
gentleman-programming
Backlog triage protocol for Engram: audit open issues and PRs, classify each item, infer maintainer ideology from comments, and produce an actionable…
SkillAFreshness: A★ 7.6KClaude Code +9 - 0.720score
engram-commit-hygiene
gentleman-programming
Commit and branch naming standards for Engram contributors, enforced by GitHub rulesets.
SkillAFreshness: A★ 7.6KClaude Code +9 - 0.720score
engram-architecture-guardrails
gentleman-programming
Architecture guardrails for Engram across local store, cloud sync, dashboard, and plugins.
SkillAFreshness: A★ 7.6KClaude Code +9 - 0.720score
engram-branch-pr
gentleman-programming
PR creation workflow for Engram following the issue-first enforcement system.
SkillAFreshness: A★ 7.6KClaude Code +9 - 0.720score
engram-business-rules
gentleman-programming
Product and business-rule guardrails for Engram.
SkillAFreshness: A★ 7.6KClaude Code +9
Cite this page
essay_crafter skill. Agent Mart. https://agentmart.dev/skills/fegome90-cmd-essay-crafter (last updated 2026-10-24)