Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add rome-os/rome --skill app_tagline_backfillgit clone --depth 1 https://github.com/rome-os/romeWrote 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/rome-os/rome/app_tagline_backfill)<a href="https://agentmods.dev/skills/rome-os/rome/app_tagline_backfill"><img src="https://agentmods.dev/badge/skills/rome-os/rome/app_tagline_backfill/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/rome-os/rome/app_tagline_backfill"><img src="https://agentmods.dev/badge/skills/rome-os/rome/app_tagline_backfill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00000 | $0.01148 |
| Opus 5.5 | $0.00000 | $0.00459 |
| Sonnet 5.5 | $0.00000 | $0.00230 |
| Haiku 4.5 | $0.00000 | $0.00115 |
Grade A, and why
app_tagline_backfill scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
App Tagline Backfill
Apps created before tagline existed share a card with an empty description.
This skill adds one sentence to each such app, commits it, and reinstalls so
the card is redrawn. It edits nothing else.
Modes
- AUTO — the caller (the coding app's
agent-turn-finishedhook) hands you the absolute source roots to process. Nobody is watching: skip the confirmation step, process every listed app, and finish with the report. - MANUAL — the guardian asked. Discover the apps yourself (step 1), show the draft table (step 3), and write only after they approve or edit it.
Step 1: Discover (MANUAL only)
LOCK="$HOME/.rome/${ROME_PROFILE:-default}/apps.lock.json"
node -e '
const lock = JSON.parse(require("fs").readFileSync(process.argv[1], "utf8"));
for (const [id, e] of Object.entries(lock.apps)) {
if (e.source?.mode === "source" && e.enabled && e.state === "installed") console.log(id, e.source.path);
}
' "$LOCK"
For each printed root, an app qualifies when <root>/app.yaml has a web:
block and no tagline: line:
grep -q '^web:' "$ROOT/app.yaml" && ! grep -q '^tagline:' "$ROOT/app.yaml" && echo "$ROOT"
Skip everything else and say why in the report: appstore entries (read-only
bundles, the author owns them), first-party apps (shipped with Rome, already
done), apps without web: (no card), apps that already have a tagline.
Step 2: Draft one sentence per app
Read name, description, and README.md (if present) under the root.
Write the tagline by the same rule app_creation/REFERENCE.md gives:
- One sentence, at most 80 characters (40 for CJK). No line breaks, no double quotes.
- Lead with what the user gets. No implementation details, jargon, or acronyms. Read it like an App Store subtitle.
- Do not copy
description— it is written for agents, not people.
Good: Track every mortgage rate that matters to you, updated daily.
Bad: A cron-driven scraper that stores rates in SQLite. (mechanism, not benefit)
Bad: Mortgage Monitor (a name, not a sentence)
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.
- 1mo ago First seen · 116 lines · 0 tokens per session scan A 73acf4cc6147
app_tagline_backfill is a skill published in the GitHub repository rome-os/rome (791 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,148 tokens. 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 2026-09-13.
Other skills, from other repositories
meta-pre-commit-quality-gate
Run three quality gates (ruff + mypy + pytest) in parallel over the staged diff, then arbitrate a single BLOCK/APPROVE verdict. Use before committing changes locally when you want a comprehensive pre-commit gate beyond per-file linting — exactly the same gate set CI enforces.
commit
Create git commits with good messages. Use when user says "commit", "create commit", or asks to commit changes.
commit
Generate git commit messages from staged changes.
knowledge-shell-process-and-worktree
Shell and Git worktree support for running commands, managing long-lived processes, and giving sessions isolated working directories. A Git worktree is a separate checkout linked to the same repository.
datalad
Retrieve, version, and publish scientific datasets with DataLad and git-annex, and capture computational provenance with datalad run, rerun, and containers-run. Use when cloning or fetching data from OpenNeuro, DANDI, datasets.datalad.org, or any DataLad dataset; when a file in a dataset reads as a broken symlink or a…
speckit-checkpoint-commit
Commit changes at meaningful checkpoints throughout the workflow.