Borrowing it
Nothing to install: this file belongs to Sma1lboy/rove. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Sma1lboy/rove/main/.claude/skills/file-issue/SKILL.mdgit clone --depth 1 https://github.com/Sma1lboy/roveWrote 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.
[](https://agentmods.dev/skills/sma1lboy/rove/file-issue)<a href="https://agentmods.dev/skills/sma1lboy/rove/file-issue"><img src="https://agentmods.dev/badge/skills/sma1lboy/rove/file-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.
<a href="https://agentmods.dev/skills/sma1lboy/rove/file-issue"><img src="https://agentmods.dev/badge/skills/sma1lboy/rove/file-issue.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 72 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00184 | $0.02347 |
| Opus 5.5 | $0.00074 | $0.00939 |
| Sonnet 5.5 | $0.00037 | $0.00469 |
| Haiku 4.5 | $0.00018 | $0.00235 |
Grade A, and why
file-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 17d 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.
How it starts
The opening of the file, as written. The whole thing — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.
File a GitHub issue
Turn a rough idea, bug report, or low-priority task into a well-structured GitHub
issue and file it with gh. The output should be something an outside contributor
can pick up cold — accurate file pointers, clear scope, explicit acceptance — not a
one-line stub.
Scope boundary (read first)
Rove tracks its own work in the daemon-owned issue store (rove api issue-*, the
web Issues page) — see docs/WORK-TRACKING.md. That
is the default for internal backlog.
This skill is for outward-facing GitHub issues — work you intend to hand to
external contributors (especially good first issues), or a public bug report. If the
user just wants to jot down internal backlog, prefer the daemon store and say so. If
they explicitly want a GitHub issue, proceed here.
Modes
The skill handles three shapes of input — detect which from the user's ask:
- Single — one idea / bug → one issue. Draft it, auto-classify (Step 1), confirm, file. The lightest path; fine to file directly once the draft + labels are shown.
- Batch — a list of ideas / a backlog dump / "file these N as issues" → many issues. Draft all, classify each, present the whole set as a table (title + proposed labels) for one confirmation, then file them in a loop and report URLs together.
- From unsolved problems — the user points at problems that came up but weren't
fixed this session ("发没解决的上去") — e.g. a known bug you hit, a deferred follow-up,
a TODO you found. Convert each into an issue: capture the symptom/repro while it's
fresh, mark it
bugorenhancement, and note in the body that it's filed-not-fixed (no implied owner). Then batch-confirm and file as above.
Pick the mode, then follow the per-issue steps below for each issue.
Hard rules (non-negotiable)
- Filing is an outward-facing publish. Show the draft(s) and labels before
gh issue create— never create silently. A batch needs one explicit confirmation; a single obvious issue can be filed right after you show it. Editing an already-filed issue is fine without re-asking. - No AI / Anthropic / Claude / Codex attribution anywhere in the title, body, or comments (per CLAUDE.md). No "Generated with…" footers.
- Never invent file paths, line numbers, flags, or function names. Every pointer in the body must be verified against the current tree (see Step 2). A wrong pointer costs a contributor 30–50% of their session reconciling a false premise.
- Default language: English for issue titles and bodies (the repo's contributor language), even when the conversation is in Chinese. Narrate to the user in whatever language they're using. If the user asks for another language, follow that.
- One concern per issue. If an idea bundles several independent tasks, split it into several issues (e.g. "add a theme" is one issue per theme).
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.
- 17d ago Changed 16b0f6ba9439
- 28d ago Changed · +1 lines 5fef63e52c06
- 1mo ago First seen · 202 lines · 184 tokens per session scan A d4b28546a142
file-issue is a skill published in the GitHub repository Sma1lboy/rove (178 stars, last pushed today), licensed MIT. It adds 184 tokens to every session and 2,347 once invoked, about $0.0007 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.
Other skills, from other repositories
fleet-orchestrator
Use when you orchestrate several workers at once (Codex or vibe sessions and Claude subagents) delivering parallel pull requests, on one machine or across several, for a human operator who does QA and merges. Covers who may write where, orchestration state that survives a reboot, restart or compaction, starting…
cross-repo-cache
Check if cross-repo coordinator results can be reused, avoiding redundant Sonnet agent launches. Compares peer repo git hashes against stored hashes from last coordinator run. Triggers on: entering Phase 1.4.1, checking cross-repo cache, before discovering peers. Returns CROSSREPOCACHEHIT with cached status or…
apex-status
CTO-level project status from git and codebase state. Use when asked "where are we", "project status", "what's done", or at the start of a work session.
tm-marathon-config-example
Example Marathon Configuration for CLAUDE.md - copy the section below into your project's CLAUDE.md.
ops-standup
Cross-repo morning briefing. Aggregation of recent commits, PRs, CI, blockers and priorities of the day. Trigger when the user wants a standup, an activity summary, or to know what happened.
work-batch
Sequential execution of user stories from a PRD file. Autonomous mode that implements and commits each story one by one. Trigger when the user wants to process a backlog, execute multiple stories, or launch an autonomous mode.