Atlas is a source-control system for coding agents that links each commit to the agent session, prompts, tool calls, and file changes that produced it. It is for developers who run multiple agents, switch between them, and need their work, decisions, and project knowledge tracked together. The catalogue includes a skill for working with Atlas.
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 pacifio/atlas --skill atlas-self-configuregit clone --depth 1 https://github.com/pacifio/atlasWrote 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/pacifio/atlas/atlas-self-configure)<a href="https://agentmods.dev/skills/pacifio/atlas/atlas-self-configure"><img src="https://agentmods.dev/badge/skills/pacifio/atlas/atlas-self-configure/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/pacifio/atlas/atlas-self-configure"><img src="https://agentmods.dev/badge/skills/pacifio/atlas/atlas-self-configure.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 Rogue Agent · line 28 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00019 | $0.01402 |
| Opus 5.5 | $0.00008 | $0.00561 |
| Sonnet 5.5 | $0.00004 | $0.00280 |
| Haiku 4.5 | $0.00002 | $0.00140 |
Grade A, and why
atlas-self-configure 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 5d 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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Atlas self-configure
Atlas's user-facing preferences live in one human-editable file,
config.toml. This skill is how you (an agent) read and change it safely —
without touching anything else Atlas persists.
Where the file lives
~/.config/atlas/config.toml
The same path on every platform. If $XDG_CONFIG_HOME is set to an absolute
path, it's $XDG_CONFIG_HOME/atlas/config.toml instead — check that variable
before assuming.
Editing that file is the supported mechanism, and the only one available to you: Atlas's config commands are IPC endpoints reachable from its own UI, not agent tools. Atlas watches the file and validates every change you make (see step 6 below), so a direct edit is a first-class way in — not a workaround.
If the file does not exist yet, Atlas hasn't created it (a fresh install before first launch, or the user deleted it). Do not create one speculatively — report that to the user instead of guessing at defaults.
One project only: the project's own config.toml
A project can override a few keys for itself in a config.toml inside the
project's .atlas/ directory. Use it only when the user asks for a change in
this project rather than everywhere. Its header says what it is; the rules:
- It holds only the keys the project changes. The allowed keys are the ones whose comment in the global file says a project can override it. Atlas ignores any other key there.
- Some of those a project can only turn off (the comment says "never on"). If the user wants one on for a single project while it's off everywhere else, say that Atlas won't honour it and stop.
- Delete a key's line to go back to the global value.
- The same care applies as below: read first, explain, change only your
lines, write atomically. It may not exist yet; unlike the global file,
creating it is fine — start it with
schemaVersion = 1and[settings].
The schema is in the file
Atlas writes a comment above every key: what it does, its default, and any constraint on its value — a numeric range, an exact set of allowed strings, "must not be empty". Read those comments and follow them. They are generated from the same table Atlas validates against, so they cannot drift from what it will actually accept.
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.
- 5d ago Changed · +17 lines 4a4e52443301
- 1mo ago First seen · 106 lines · 19 tokens per session scan A 99804208a1d2
atlas-self-configure is a skill published in the GitHub repository pacifio/atlas (9,549 stars, last pushed today), licensed Apache-2.0. It adds 19 tokens to every session and 1,402 once invoked, about $0.0001 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 2026-09-03.
Other skills, from other repositories
aidlc-portfolio
Coordinate multiple AI-DLC workflows across repositories and Git worktrees using an evidence-backed portfolio catalog and deterministic workspace tooling. Use when initializing an AI-DLC portfolio workspace, discovering organization or business context, registering projects and dependencies, creating child intents and…
git-ai-search
Search and restore AI conversation context from git history.
mcore-split-pr
Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.
agent-deck
Helper scripts live in this skill's scripts/ folder. Resolve them from the base directory shown when the skill loaded, not from the user's project.
machinist
Use Machinist to create, assign, and monitor software tasks. Use when a coding agent needs to work with Machinist, its GitHub issue workflow, intake labels, direct runs, or managed queue.
pr-explainer
Use when a developer asks to explain a pushed PR, walk through a PR stack, understand one stack member, or explain what an AI agent just pushed so they can connect code changes to end-to-end functionality and update domain knowledge. Not for defect review, merge approval, PR-body authoring, or full-repository…