Journal House Style: Fit Any Venue's Author Guidelines

What it does

Editors and thesis offices check the shape of a manuscript before anyone reads it: the word limits, the sections they require, the statements they demand, how figures are labelled and where captions go. Every venue sets these differently, and the rules live on pages that change. This skill makes your agent fit your manuscript to the venue you name, from that venue's own current guidelines, or to the requirements you give it.

• A venue profile built from the venue's own pages. Name a journal, conference, publisher series or thesis office. The agent finds its current author guidelines, using a bundled guide to where about twenty publishers, the major ML, vision and language conferences, LaTeX template galleries and university thesis offices keep them. It fetches the pages and writes a profile for your article type: word and character limits and what the count excludes, display items and references, section order, required sections and statements, heading depth, caption rules, reference style and template.
• Or your own requirements. A lab style sheet, a department thesis guide, a template or an editor's email becomes the same profile. This route works offline.
• A source on every rule, nothing from memory. Each rule records the page it came from and the date it was read (or "user-supplied"). What the sources do not say is recorded as "not stated", never filled in with a typical value. When two pages disagree, both are recorded and you are told.
• An audit you can act on. One table of what is over a limit, missing, forbidden or out of order, each finding citing the rule it breaks, then a prioritised list: missing required content first, then structure, cuts and format.
• Restructure, cut, recheck. Sections reordered, required statements added, captions relabelled and moved, material moved rather than deleted, then cut to budget and audited again. Anything only you can supply (a DOI, an ethics approval, a grant number) becomes a visible placeholder.

The audit script

venue_audit.py (Python 3.10+, standard library, no network) validates the profile and audits a Markdown or LaTeX draft, one file or a folder of section files. It checks every rule it can count: title, abstract and per-section word limits, the main-text limit with the components the venue counts, captions and caption titles, figures, tables and references, heading depth, section order, required statements, forbidden phrases, citations or headings in the abstract, subheading rules, caption label and placement. Every other rule in the profile is listed for the agent to check by reading. In the demo it flagged a 323-word abstract against a 300-word limit, a missing Author Summary and Data Availability Statement, "data not shown", a citation and headings in the abstract, "Figure 1" callouts where the journal uses "Fig 1", and captions in the wrong place.

What's included

• SKILL.md: the five-step workflow (profile, audit, restructure, cut, check) and its rules
• references/where-to-look.md: where house styles live (publishers, conferences, template galleries, thesis offices), each pattern checked, and how to fetch pages that block automated tools
• references/profile-format.md: the profile format, every key, and the checklist of what to extract
• references/generic-structure.md: generic research-article heuristics, labelled as such
• scripts/venue_audit.py and examples/: a profile built from fictional thesis-chapter requirements, a draft and its audit

Part of the Aurora Lab research pipeline

Works entirely on its own. With a research/ folder it writes research/venue-profile.md and research/paper/venue-audit.md; without one, it works next to your draft. Pairs well with Aurora Lab's Scientific Voice (how the sentences read) and Claim Check (does each sentence stay inside its evidence?), but needs neither.