Agent skill

File Issue

by rome-os in rome-os/rome

File one GitHub issue from a description the user gives — classify it as a bug report, feature request, or task spec, gather what the body needs from the tracker and the code, ask the user only for…

MITAuto-check passedTesting & QA

Install File Issue

skills CLI
$ npx skills add rome-os/rome --skill file-issue -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install rome-os/rome file-issue --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/rome-os/rome.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/file-issue .claude/skills/file-issue && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
file-issue
GitHub stars
737
Token cost
~1.7k tokens
SKILL.md length
981 words
Files
1
Skills in repo
18
Repo updated
First seen
Licence
MIT

At a glance

File one GitHub issue from a description the user gives — classify it as a bug report, feature request, or task spec, gather what the body needs from the tracker and the code, ask the user only for…

  • Works in 7 steps: Classify → Gather → Ask → …
  • The user says file an issue / bug / feature request / task for <description
  • SKILL.md covers 1. Classify, 2. Gather, 3. Ask and 4. Draft, plus 3 more sections
  • Calls gh; reaches github.com

What it does

File Issue is an agent skill from rome-os/rome. File one GitHub issue from a description the user gives — classify it as a bug report, feature request, or task spec, gather what the body needs from the tracker and the code, ask the user only for what those cannot answer, draft it in the repo's issue format, and create it with gh after a go-ahead. Use when the user says "file an issue / bug / feature request / task for <description".

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Testing & QA, covering QA and bug reports. It works with GitHub. The repository describes itself as: A compounding agent OS for recursive agents. Also an open source alternative to Grok Bot and Meta's Muse. The licence is MIT.

When your agent uses it

  • The user says file an issue / bug / feature request / task for <description
  • Tasks that involve QA and bug reports

Example prompts

  • “file an issue / bug / feature request / task for <description”
  • “/file-issue”

Workflow steps

7 steps, taken from the step headings in SKILL.md.

  1. Classify
  2. Gather
  3. Ask
  4. Draft
  5. Approve
  6. Create
  7. Verify

What it can do on your machine

Read from SKILL.md and the folder at commit f187f45. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • gh

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • github.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

File Issue loads about 1.7k tokens when it runs. Until then it costs about 101 tokens; SKILL.md has 981 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~101
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from rome-os/rome at commit f187f45, republished under its MIT licence (© rome-os). 981 words, ~1,709 tokens.

Download SKILL.mdSave it as .claude/skills/file-issue/SKILL.md (or your agent's skills folder).
name
file-issue
description
File one GitHub issue from a description the user gives — classify it as a bug report, feature request, or task spec, gather what the body needs from the tracker and the code, ask the user only for what those cannot answer, draft it in the repo's issue format, and create it with `gh` after a go-ahead. Use when the user says "file an issue / bug / feature request / task for <description>".
argument-hint
description
disable-model-invocation
true

File an issue

Goal end-state: one issue on the tracker whose type, title, body sections, and labels follow docs/authoring/github-issues.md and the rulebook of its type, with every required section filled from evidence.

The description arrives as the skill argument or in the conversation: $ARGUMENTS

The authoring docs own the issue format, and a rule there wins over any wording here. Read the shared rules first, then the rulebook of the type chosen in step 1:

Prose in the title and body follows docs/authoring/WRITING.md. Situation uses the terms in docs/concepts/, and a term links to its entry on first use with a full https://github.com/<owner>/<repo>/blob/main/docs/concepts/<file>.md#<anchor> URL, because a relative link does not resolve inside an issue body.

1. Classify

Decide the type with the two questions from github-issues.md, in order:

  1. Does the system break its own intended behavior? → bug report.
  2. Is the change already decided and scoped? → task spec.
  3. Otherwise → feature request. A pain point with no feature attached, and a feature idea, both file as a feature request.

When the description supports two readings that change the type — a symptom the user may or may not consider intended behavior, or a change the user may or may not have decided — ask which one holds. Carry the question into step 3 rather than asking it alone.

2. Gather

Fill the required sections of the chosen type from three sources, in this order: the description, the tracker, the code. When step 1 left two readings open, gather what both readings need, so the answer in step 3 picks a draft rather than a second gather.

  1. Resolve the repo with gh repo view --json nameWithOwner. The concept URLs in Situation need the owner and repo name.
  2. Search the tracker with gh issue list --search "<keywords>" --state all. If an open issue already tracks the same defect, gap, or change, report it and stop. If a closed one covers it, link it from the body.
  3. Read the code and the docs that the description touches. For a bug report, reproduce the symptom when the steps are cheap, and gather the evidence Initial triage and Suspected root cause need. For a feature request, find the PRs, incidents, or threads where the gap bit, because Pain carries evidence. For a task spec, read the files Scope names, and decide for each Acceptance item whether a committed test or a one-time check proves it.
  4. For a bug report, pick one of P0, P1, P2, P3 from the label descriptions in gh label list.
  5. For a task spec, find the issues it depends on and record them as a Blocked by line by number.

Step 3 answers what the code can answer. A question goes to the user only for what the description, the tracker, and the code leave open.

3. Ask

Collect every open question from steps 1 and 2 and ask them in one round, with one AskUserQuestion call. A question earns a slot when a required section cannot be written without the answer, or when the answer changes the type. Every question offers the readings the evidence supports as options, with the most likely one first.

Questions that earn a slot:

  • the type, when step 1 left two readings open.
  • what the user expected to happen, when the bug rulebook's Symptom needs it and the description states only what happened.
  • who pays the pain and where it bit, when the feature rulebook's Pain has no evidence to point to.
  • what stays out, when the task rulebook's Scope has no boundary.
  • the priority of a bug, when the evidence supports two labels.

Skip the round when no question earns a slot. Ask a second round only when an answer opens a gap the first round could not foresee.

Show full SKILL.md (352 more words)Show less

4. Draft

  1. Write the title in the shape github-issues.md fixes and in the mood the type's rulebook fixes.
  2. Write every required section of the type, in the order its rulebook lists, and only the optional sections the evidence fills. Situation opens the body and names no file, symbol, or line number. Sections after Situation carry the exact strings, paths, and commands.
  3. For a feature request, write Possible directions only when the user presents a direction in the description or in an answer, and list only the directions the user presented. Otherwise omit the section. A direction the skill would invent on its own stays out of the issue.
  4. Set the labels: the type label, plus exactly one priority label for a bug report, plus ready-for-agent for a task spec an agent can implement from the body alone. A feature request never carries ready-for-agent.
  5. Check the draft against the type's rulebook once more: every required section present, Possible fixes and Possible directions pick nothing, Suspected root cause states its confidence, every Acceptance item names what proves it, Scope names what stays out.

5. Approve

Show the full draft — title, labels, body — and ask for the go-ahead in one line. Write nothing to GitHub before the user gives it. Apply the user's edits to the draft and show the changed part again before filing.

6. Create

  1. If gh label list lacks a label the draft carries, stop and report the missing label. The repo's label list is the only source of label descriptions, so the skill creates no label.

  2. Create the issue, passing the body through a file so that backticks and quotes survive the shell:

    bash
    gh issue create --title "<title>" --body-file <path> --label <type-label> [--label <priority-or-ready-for-agent>]

7. Verify

  1. Read the issue back with gh issue view <number> --json title,labels,body and confirm the title, every label, and every required section landed.
  2. For a task spec with a Blocked by line, confirm each number names an open issue with gh issue view <blocking-number> --json state.
  3. Report the issue URL, its type, and its labels.

© rome-os, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/file-issue of rome-os/rome.

Open the folder on GitHubat commit f187f45

Compare with similar skills

File Issue next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

File Issue compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
File Issue this skillrome-os/rome737—~1.7kAutomated safety check: PassMIT
Weavebench Cua ReproduceAMAP-ML/LongHorizon-Harness1.7k—~1.6kAutomated safety check: PassMIT
Evidence-Driven Testingmichaelshimeles/skills1.3k1 repos~3.9kAutomated safety check: PassNone
Create GitHub IssueNVIDIA/OpenShell16k—~1.7kAutomated safety check: PassApache-2.0
Triage IssuesClickHouse/clickhouse-java1.6k—~904Automated safety check: PassApache-2.0
Gentle AI Issue CreationGentleman-Programming/gentle-shell1.2k—~2.5kAutomated safety check: PassApache-2.0

Similar skills

  • Weavebench Cua Reproduce

    AMAP-ML/LongHorizon-Harness

    Reproduce CUA-Harness experiments on WeaveBench from a GitHub checkout.

    1.7k GitHub stars~1.6k tokensUpdated 1 mo ago
    Testing & QAAuto-check passed
  • Evidence-Driven Testing

    michaelshimeles/skills

    Records an annotated screen recording of the agent testing an app hands-on, then posts the video and a results summary to the PR and tracker issue.

    1.3k GitHub starsUsed in 1 repo~3.9k tokens
    Testing & QAAuto-check passed
  • Create GitHub Issue

    NVIDIA/OpenShell

    Official

    Create GitHub issues using the gh CLI. An agent skill from NVIDIA/OpenShell.

    16k GitHub stars~1.7k tokensUpdated today
    Testing & QAAuto-check passed
  • Triage Issues

    ClickHouse/clickhouse-java

    Analyzes a single GitHub issue at a time. An agent skill from ClickHouse/clickhouse-java.

    1.6k GitHub stars~904 tokensUpdated today
    Testing & QAAuto-check passed
  • Gentle AI Issue Creation

    Gentleman-Programming/gentle-shell

    Create and triage GitHub issues from repository evidence. An agent skill from Gentleman-Programming/gentle-shell.

    1.2k GitHub stars~2.5k tokensUpdated today
    Testing & QAAuto-check passed
  • Termio Bug Report

    termio-sh/termio

    Diagnose a termio hang, beachball, crash, or 'it froze again' from the evidence macOS and termio actually leave behind — live process samples, crash and CPU-burn reports, the unified log, the…

    537 GitHub stars~1.9k tokensUpdated today
    Testing & QAAuto-check passed

More from rome-os/rome

All 18 skills in this repo
  • Color Audit

    rome-os/rome

    Audit a design system's color palette against measurable color-science disciplines — WCAG/APCA contrast of declared token pairs, perceptual (OKLCH) ramp uniformity, color-blindness safety of…

    737 GitHub stars~2.7k tokensUpdated today
    Auto-check passed
  • Audit the subordinate copy in a UI — section descriptions, field helper text, hints, card subtitles, empty-state body copy, tooltip bodies — against the secondary-text ruleset, and emit a per-string…

    737 GitHub stars~2.8k tokensUpdated today
    Auto-check passed
  • UX Semantics Audit

    rome-os/rome

    Audit an existing UI/UX design (React/JSX/TSX components, HTML, or generated app code) against a tiered ruleset of verifiable UX principles, and produce structured, evidence-cited findings that a…

    737 GitHub stars~2.1k tokensUpdated today
    Auto-check passed
  • Add a new Rome-managed OAuth integration for a third-party service so a user can delegate access by clicking Connect, and Rome can act on the service with the delegated token (the GitHub/Slack model…

    737 GitHub stars~3.6k tokensUpdated today
    Auto-check passed
  • Composio CLI

    rome-os/rome

    Help users operate the published Composio CLI to find the right tool, connect accounts, inspect schemas, execute tools, subscribe to trigger events with composio listen, script workflows with…

    737 GitHub stars~2.7k tokensUpdated today
    Auto-check passed
  • Routine From Chat

    rome-os/rome

    Turn a plain-language "do this automatically" request into a routine — either event-triggered ("text me when I get an email from my landlord") or scheduled ("remind me every Friday at 9am").

    737 GitHub stars~3.8k tokensUpdated today
    Auto-check passed

Works with

Categories

Questions about File Issue

What does File Issue do?

File one GitHub issue from a description the user gives — classify it as a bug report, feature request, or task spec, gather what the body needs from the tracker and the code, ask the user only for…. File Issue is an agent skill from rome-os/rome. File one GitHub issue from a description the user gives — classify it as a bug report, feature request, or task spec, gather what the body needs from the tracker and the code, ask the user only for what those cannot answer, draft it in the repo's issue format, and create it with gh after a go-ahead.

When should I use File Issue?

File Issue fits situations like: the user says file an issue / bug / feature request / task for <description; tasks that involve QA and bug reports.

How do I install File Issue in Claude Code?

Run `npx skills add rome-os/rome --skill file-issue -a claude-code`. Or copy the skill folder (.claude/skills/file-issue in rome-os/rome) into .claude/skills/file-issue in your project. Claude Code loads it when a task matches its description.

How do I install File Issue in Codex?

Run `npx skills add rome-os/rome --skill file-issue -a codex`. Or copy the skill folder (.claude/skills/file-issue in rome-os/rome) into .agents/skills/file-issue in your project. Codex loads it when a task matches its description.

Can I use File Issue in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add rome-os/rome --skill file-issue -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/file-issue, .gemini/skills/file-issue, .github/skills/file-issue and .opencode/skills/file-issue in your project.

What does File Issue need to run?

Going by SKILL.md and its folder, File Issue needs the command-line tools its instructions call (gh).

Does File Issue access the network?

SKILL.md names 1 domain. In commands or code: github.com; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is File Issue safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does File Issue use?

File Issue is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does File Issue use?

About 1.7k tokens (SKILL.md is roughly 6.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to File Issue?

Skills that share tags, products or a category with File Issue: Weavebench Cua Reproduce (AMAP-ML/LongHorizon-Harness, 1.7k stars), Evidence-Driven Testing (michaelshimeles/skills, 1.3k stars), Create GitHub Issue (NVIDIA/OpenShell, 16k stars) and Triage Issues (ClickHouse/clickhouse-java, 1.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains File Issue?

rome-os (a GitHub organization) maintains it in rome-os/rome, which has 737 GitHub stars. The repository holds 18 skills in this directory. The repository was last updated on October 9, 2026.

Source: rome-os/rome on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.