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 fledgeling-co/fledgeling-plugins --skill atlas-publishgit clone --depth 1 https://github.com/fledgeling-co/fledgeling-pluginsWrote 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/fledgeling-co/fledgeling-plugins/atlas-publish)<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/atlas-publish"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/atlas-publish/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/fledgeling-co/fledgeling-plugins/atlas-publish"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/atlas-publish.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 | $0.00281 | $0.04710 |
| Opus 5 | $0.00140 | $0.02355 |
| Sonnet 5 | $0.00056 | $0.00942 |
| Haiku 4.5 | $0.00028 | $0.00471 |
Grade A, and why
atlas-publish scanned grade A with 1 finding 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 4d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| Blob upload | `curl -sSI <returned blob url>` | not `200`, or `content-length` differs from the local file's byte count | How it starts
The opening of the file, as written. The whole thing — 234 lines — stays where its author put it; the contents beside it link to each section on GitHub.
atlas-publish — Atlas release conductor
You take the currently-open PRs in the Atlas monorepo and drive them to a registered draft release — either an over-the-air (OTA) JavaScript bundle or an App Store binary. The heavy work is real: you merge branches, archive a binary, upload it to TestFlight, export a JS bundle, put bytes in a public Blob store, and write release rows to production Mongo. The one thing you do not do is make any of it reach an end user.
You run in the founder's Claude Code on their Mac, against production infrastructure.
Running as a Gemini model? Read gemini.md in this directory first, then follow this file with the overrides it names. It turns the zero-publish rule, the Step 0 gate states, the per-asset upload denominator and the review-per-PR count into ledgers with read-back commands, and names which of Steps 3 and 5 to hand to another model. Other models skip it.
Why draft is where automation stops
Not because the step is hard, and not because you might get it wrong. A draft OTA bundle and a TestFlight build are re-buildable artifacts — a bad one is deleted, re-exported and registered again inside one export cycle. A published bundle is not: it lands on phones you do not control, and retracting it changes what the next boot serves rather than undoing the one that already happened. Tolerance tracks how reproducible the target is, not how good the checking is, so no amount of extra checking moves a customer-facing publish inside the envelope.
Concretely, this skill calls register_bundle and create_app_version, each with confirm: true,
and each only ever writes status: draft. It does not call publish_bundle, publish_app_version,
retract_bundle, retract_app_version or set_min_app_version. When you reach the end of a run,
hand the founder the three publish routes (references/handoff.md) and stop.
The boundary is restated at Step 9 and Step 11, where the temptation actually sits — a rule stated once in frontmatter and needed nine steps later is a rule the run has already forgotten. Measured reason: instruction-following degrades 39% on average from single-turn to multi-turn across 15 models, which is why the gates below are exit codes rather than paragraphs.
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 4d ago Changed · +1 tokens per session 0dc82eed1a05
- 12d ago First seen · 234 lines · 280 tokens per session scan A 04874dd9a3a3
atlas-publish is a skill published in the GitHub repository fledgeling-co/fledgeling-plugins (2 stars, last pushed 3d ago), licensed MIT. It adds 281 tokens to every session and 4,710 once invoked, about $0.0014 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
agent-release-swarm
Agent skill for release-swarm - invoke with $agent-release-swarm.
release-announcement
Write a release announcement — changelog, blog post, in-app note, or social post — that leads with user impact, names the audience, and includes upgrade/migration steps without filler.
multi-agent-release-manager
Cleans up the workspace, formats code, runs presubmit checks, and uploads CLs to Gerrit.
release-notes
Generate user-facing release notes from tickets, PRDs, or changelogs. Creates clear, engaging summaries organized by category (new features, improvements, fixes). Use when writing release notes, creating changelogs, announcing product updates, or summarizing what shipped.
pack-submit
Package one of this agent's own skills as a standalone community pack and submit it to the aeon registry as a PR.
updater_guide
Guidance for checking for and installing Row-Bot updates.