agentleFS
gigio1023gigio1023/agent-skills/skills/additional/credo-release/SKILL.md

credo-release

Skill2 starsChanged 9 days agoMITView on GitHub

Check a LaTeX paper before submission, camera-ready, or arXiv release: run bundled source checks, inspect the rendered artifact and current venue requirements, and ask for author-only attestations. Use for final checks or linting TeX. NOT for argument planning (credo-paper-plan), a LaTeX report or result bound for a customer or other external reader (evaluation-audit release mode), automatic paper rewriting, or submission without authorization.

1 other project uses a version of this.

Adopted by

Most recently taken up by

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:

    Keep new recurring mistakes in a project checklist when that maintenance is authorized.

    line 31

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Credo: Release

Return a release review with evidence for checked items and the remaining author decisions. Adapted from Carlini's Research Paper Release Checklist; see coverage for the bundled script's limits.

Scope: a paper or preprint headed to a venue or arXiv. A customer, technical, or evaluation report typeset in LaTeX goes to evaluation-audit release mode, which decides whether its results can be shown; the source scanner below can still lint that report's TeX when asked.

Source checks

From this skill directory:

python3 scripts/check_tex.py path/to/main.tex --log path/to/main.log
python3 scripts/check_tex.py path/to/main.tex --blind --author "Full Name"

Python 3.10 or newer; no packages. The script follows \input and \include, strips comments, and prints severity, location, and message plus a summary. Exit 1 means an error finding; 2 means invalid usage. It catches source markers, potential text mistakes, selected anonymity clues, and supplied build-log issues. Warnings need interpretation; it cannot establish venue compliance or scientific correctness.

Report the command and actionable findings. Rerun affected checks after authorized edits. If Python is unavailable, inspect by the alternatives in the coverage reference and state the narrower coverage.

Artifact and author checks

Use the release checklist for the actual release type. Inspect available artifacts directly with suitable tools: PDF rendering, page count, fonts, captions, links, title/form comparison, and applicable venue rules. Check the current official requirements when needed; a remembered page limit or anonymous-author format is insufficient.

Record checked only with the actual observation, confirmed by author only for an author answer, open for an unresolved item, and n/a with the applicable reason. Do not infer rendered layout from source or claim another-machine verification from one local render. Ask only for items requiring unavailable material or human attestation, such as agreement to authorship or the final submission declaration; bundle related missing answers rather than walking the user through checks already completed.

Keep new recurring mistakes in a project checklist when that maintenance is authorized. Do not silently edit an installed shared package from a paper's release review.

Output and authority

Lead with release-blocking findings, then the compact checked/open record needed by the author. Distinguish an actual defect from unavailable evidence. A review does not authorize rewriting, uploading, submitting, or certifying on the author's behalf. If fixes or submission were already requested, preserve that authority and complete only the requested actions; structural argument work can use credo-paper-plan.

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.