What is resolve-issue?

A workflow for taking a public GitHub issue from investigation to a completed fix, feature, or documentation change on its own branch. TDD means writing a failing test that demonstrates the problem before implementing the fix.

Use it to classify issues, create minimal reproductions, write tests, implement changes, and prepare a focused pull request.

It gives issue work a repeatable path and helps preserve matching behavior between a Perl implementation and its converted version.

How dginev/latexml-oxide configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to dginev/latexml-oxide. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/dginev/latexml-oxide/main/.claude/skills/resolve-issue/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/dginev/latexml-oxide

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/dginev/latexml-oxide/resolve-issue/github.svg)](https://agentmods.dev/skills/dginev/latexml-oxide/resolve-issue)

Your own site

<a href="https://agentmods.dev/skills/dginev/latexml-oxide/resolve-issue"><img src="https://agentmods.dev/badge/skills/dginev/latexml-oxide/resolve-issue/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/dginev/latexml-oxide/resolve-issue"><img src="https://agentmods.dev/badge/skills/dginev/latexml-oxide/resolve-issue.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 129 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,780 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00129	$0.03780
Opus 5.5	$0.00052	$0.01512
Sonnet 5.5	$0.00026	$0.00756
Haiku 4.5	$0.00013	$0.00378

Measured 2mo ago against content hash cc811ca171cb, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

resolve-issue scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/resolve-issue/SKILL.md · 235 lines

How it starts

The opening of the file, as written. The whole thing — 235 lines — stays where its author put it; the contents beside it link to each section on GitHub.

The repo is public; issues now arrive from real users. This skill is the outer loop that wraps the others (canvas-triage → min-repro → perl-port) into one issue-to-PR procedure. Faithful translation still governs everything: Perl is ground truth, never downgrade errors to "pass", diverge only when OXIDIZED_DESIGN.md sanctions it.

Step 0 — Branch for the ticket (first, always)

One ticket = one branch = one PR. Never work on main, and never pile a new ticket onto another ticket's branch/PR. Read the issue title to get the number + a slug, then cut a branch off an up-to-date main:

git switch main && git pull --ff-only
git switch -c fix-<N>-<short-slug>      # e.g. fix-293-subimport-standalone

Use a fix- / feat- / docs- prefix per the issue type (Step 1). If triage (Step 2) later proves the issue is parity / out-of-scope / not-ours, abandon the branch (git switch main && git branch -D fix-<N>-…) — no PR. Cut the branch off main even when another ticket's PR is open, so the two stay independent.

Step 1 — Classify the issue

Read the issue in full (gh issue view <N> --json title,body,labels,author). Pick exactly one type; it selects the path:

Type	Signal	Path
1 — Bug	"X is wrong / crashes / differs from LaTeX"	red/green TDD → patch (the main path below)
2 — Feature	"add / support / it would be nice if"	design plan first → then TDD-implement
3 — Docs	"document / explain / unclear"	refactor the doc with more explanation; verify claims against real behavior
4 — Other	meta, build, question, out-of-scope	triage; usually reduces to 1–3, else answer/ask the user

Most issues name a TeX source and its conversion → Steps 2–8 apply, PR in Step 9. Types 2/3 add a front-half; see "Type-specific notes".

Step 2 — Reproduce and confirm it's ours

Put the issue's MWE verbatim into a .tex file (scratchpad first). Convert with the current binary: cargo run --bin latexml_oxide -- --dest=/tmp/r.html repro.tex. Confirm you see the reported symptom, not something adjacent.
Classify vs Perl before writing any fix (canvas-triage golden rules): run same-host Perl verbose, never --quiet, ANSI-strip before grepping. A bug the user reports is only a straight patch when it is GENUINE-RUST-ONLY (Perl behaves correctly, Rust doesn't). If Perl misbehaves identically it is parity/upstream (KNOWN_PERL_ERRORS.md) or a surpass-perl decision — not a silent divergence. If the paper is large, shrink it with min-repro first.

Read the full file on GitHub · 235 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 235 lines · 129 tokens per session scan A cc811ca171cb

Subscribe to this mod's changes

resolve-issue is a skill published in the GitHub repository dginev/latexml-oxide (18 stars, last pushed yesterday), licensed CC0-1.0. It adds 129 tokens to every session and 3,780 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Commits Git and workflow LaTeX Planning epub html jats math-parsing mathml

resolve-issue next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
resolve-issuedginev/latexml-oxide	Skill	18	yesterday	CC0-1.0	A	129	Claude Code
pre-commit-auditflonat/flonat-research	Skill	149	3d ago	MIT	A	100	Claude Code
latex-diffflonat/flonat-research	Skill	149	3d ago	MIT	A	56	Claude Code
pro-gitmichalstrnadel/source-to-skill	Skill	8	11d ago	MIT	A	122	Claude Code
inclusio-publishersebastienrousseau/inclusio	Skill	0	2d ago	Apache-2.0	A	30	Claude Code, Codex
create-prprisma/orm	Skill	48k	yesterday	Apache-2.0	A	48	Claude Code, Codex
repo-syncdavidondrej/skills	Skill	4.3k	3d ago	MIT	A	25	Claude Code, Codex

latexml-oxide: Skill for Claude Code