Skip to content
cccskills
FreePublished on GitHub

auto-paper

Internal Harness instruction source for auto-paper. Route through visible Harness aliases instead of invoking directly.

The English summary is being prepared. Showing the original description.

See how to install

Files included (55)

  • SKILL.md4.9 KB
  • agents/openai.yaml258 B
  • assets/templates/audit_report.md182 B
  • assets/templates/caption_claim_map.md169 B
  • assets/templates/citation_audit_report.md119 B
  • assets/templates/citation_plan.md129 B
  • assets/templates/citation_support_bank.md283 B
  • assets/templates/claim_citation_map.md124 B
  • assets/templates/claim_register.md185 B
  • assets/templates/claims_to_avoid.md103 B
  • assets/templates/compile_report.md140 B
  • assets/templates/config.yaml536 B
  • assets/templates/confirmed_motivation.md260 B
  • assets/templates/exemplar_learning_dossier.md151 B
  • assets/templates/experiment_source_map.md184 B
  • assets/templates/figure_asset_map.md158 B
  • assets/templates/figure_backend_report.md165 B
  • assets/templates/figure_contract.md240 B
  • assets/templates/figure_requirement_scan.md213 B
  • assets/templates/final_gate_ledger.md184 B
  • assets/templates/intake_report.md183 B
  • assets/templates/latex_patch_plan.md161 B
  • assets/templates/latex_patch.diff1 B
  • assets/templates/logic_transfer_audit.md140 B
  • assets/templates/motivation_surface_map.md124 B
  • assets/templates/original_logic_map.md147 B
  • assets/templates/patch_ledger.md246 B
  • assets/templates/research_dossier.md117 B
  • assets/templates/revision_audit_report.md103 B
  • assets/templates/run_request_register.json78 B
  • assets/templates/run_request_register.md250 B
  • assets/templates/section_blueprints.md143 B
  • assets/templates/sota_gap_map.md114 B
  • assets/templates/source_index.md170 B
  • assets/templates/style_profile.md94 B
  • assets/templates/tex_inventory.json253 B
  • assets/templates/writing_rationale_matrix.md301 B
  • references/artifact-contract.md3.0 KB
  • references/auto-iterate-boundary.md1.7 KB
  • references/auto-paper-loop.md2.6 KB
  • references/citation-support-bank.md1013 B
  • references/latex-source-control.md1.3 KB
  • references/motivation-thread.md967 B
  • references/writing-rationale-matrix.md881 B
  • scripts/artifact_check.py606 B
  • scripts/citation_bank_check.py9.0 KB
  • scripts/claim_register_check.py8.2 KB
  • scripts/figure_requirement_scan.py8.7 KB
  • scripts/init_artifacts.py2.4 KB
  • scripts/integrity_audit.py7.8 KB
  • scripts/latex_guard.py16.3 KB
  • scripts/reference_inventory.py7.0 KB
  • scripts/revision_audit.py10.2 KB
  • scripts/style_metrics.py7.3 KB
  • scripts/tex_inventory.py7.8 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Auto Paper

Use this as the auto-paper orchestrator, not a polishing tool. It routes phases and checks artifacts without inventing paper state.

Boundaries

  • Author evidence outranks exemplars: local results, figures, tables, notes, approved claim boundaries, and operator facts.
  • Do not invent experiments, results, mechanisms, statistics, novelty, clinical meaning, data availability, citations, or source support.
  • Blogs, reviews, surveys, and tutorials are citation-supported by default unless the operator explicitly asks for an uncited opinion memo.
  • External papers are examples for structure, style, terminology, and positioning, not author claims.
  • Patch LaTeX only from written artifacts.
  • Read experiment results primarily through docs/30_evidence/Experiment_Evidence_Index.{json,md} and artifact paths named there. Direct iteration_log.json reads are weak intent signals; paper claims require reports, configs, logs, metrics, or run artifacts.
  • Read docs/context/experiments.md when present for Research Wiki context, stable observations, run requests, assurance gaps, and open questions. Legacy docs/45_discoveries/Research_Wiki.md is a fallback before dynamic-context-v2 migration; neither file is approval.
  • Do not write iteration_log.json, .auto_iterate/**, or hidden controller state. Durable paper artifacts live under auto_paper_output/<paper_id>/.

Required Config

Before the loop, establish paper_id, workflow, venue/target, draft path, TeX roots, bibliography paths, figures, materials, references, artifact dir, optional experiment evidence index, output language, citation target count, figure/table requirement status, compile command, human gate policy, and forbidden directions in auto_paper_output/<paper_id>/config.yaml or an intake artifact. If missing, route to $auto-paper-intake or return USER_GATE.

After Grill establishes an Automation Policy, paper phases auto-proceed within that policy. Use USER_GATE only for missing operator intent, explicit approval tools, or irreversible external submit.

Use the deterministic checker when the operator requests an automatic paper chain:

python tooling/auto_paper/scripts/auto_paper/artifact_check.py \
  auto_paper_output/<paper_id> --phase <phase> --json

The checker reports artifact readiness only; it does not invent manuscript content or prove scientific claims.

Phase Route

Run in order unless an audit routes back: $auto-paper-research, $auto-paper-argument, $auto-paper-citation, $auto-paper-layout, $auto-paper-patch, $auto-paper-harden.

Optional branches: $auto-paper-response, $auto-paper-data, $auto-paper-figure. Enter $auto-paper-figure when the operator asks for figure work, TeX or source files mention needed figures/tables, or harden finds figure/caption risk.

Artifacts

Durable decisions live under auto_paper_output/<paper_id>/.

  • Intake: config.yaml, source_index.md, tex_inventory.json, intake_report.md, experiment_source_map.md, optional figure_requirement_scan.md
  • Phase artifacts: research dossier, style/gap maps, confirmed motivation, claim register, citation support bank, section blueprints, rationale matrix, citation and patch plans, figure contracts, patch ledgers, audits, and final gate ledger
  • run_request_register.{json,md} when missing experiments are needed

Use references/artifact-contract.md for identifiers and stale-artifact policy.

Gate And Routing

Do not patch around missing upstream artifacts. Every audit finding names root_cause, owning_phase, required_artifact, fix_action, and downstream_risk.

Decisions: NEXT_UNIT, REWORK_RESEARCH, REWORK_ARGUMENT, REWORK_CITATION, REWORK_LAYOUT, REWORK_PATCH, RUN_REQUEST, USER_GATE, COMPLETE, ABORT.

RUN_REQUEST means writing found missing experiment evidence for $run. Record the blocking claim, needed evidence, minimum artifacts, suggested prompt, and acceptance check in run_request_register.{json,md}.

When a draft introduces, narrows, or removes a result claim, record Claim Delta Evidence in the phase artifact or Gate ledger. Unsupported claims become RUN_REQUEST, not a default approval pause.

References

Load only active-phase references from references/: loop, auto-iterate boundary, artifact contract, writing rationale, motivation thread, citation support, and LaTeX source control.

Use deterministic helpers when available:

  • .agents/skills/auto-paper/scripts/figure_requirement_scan.py
  • ../../../.agents/references/research-supervision-patterns.md
  • ../../../.agents/references/research-supervision/README.md

Output

Final responses report artifact manifest, modified files, guard/compile results, remaining reviewer risks, and next human gates.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Convert run artifacts into calibrated interpretation and next decisions.

The English summary is being prepared. Showing the original description.

linzhe001/Harness-Research87Updated Aug 9, 2026

Internal Harness instruction source for auto-paper-argument. Route through visible Harness aliases instead of invoking directly.

The English summary is being prepared. Showing the original description.

linzhe001/Harness-Research87Updated Aug 9, 2026

Internal Harness instruction source for auto-paper-citation. Route through visible Harness aliases instead of invoking directly.

The English summary is being prepared. Showing the original description.

linzhe001/Harness-Research87Updated Aug 9, 2026

Internal Harness instruction source for auto-paper-data. Route through visible Harness aliases instead of invoking directly.

The English summary is being prepared. Showing the original description.

linzhe001/Harness-Research87Updated Aug 9, 2026

Internal Harness instruction source for auto-paper-figure. Route through visible Harness aliases instead of invoking directly.

The English summary is being prepared. Showing the original description.

linzhe001/Harness-Research87Updated Aug 9, 2026

Internal Harness instruction source for auto-paper-harden. Route through visible Harness aliases instead of invoking directly.

The English summary is being prepared. Showing the original description.

linzhe001/Harness-Research87Updated Aug 9, 2026

See all skills by linzhe001

Report a problem with this skill