egregore: Skill for Claude Code

.claude/skills/onboarding/SKILL.md

onboarding is a skill for Claude Code from egregore-labs/egregore. It costs 38 tokens per session (5,604 once invoked), scanned A, original, MIT.

A welcome process for a new Egregore member that verifies access, explains how the system works, invites collaboration, and guides their first handoff. A handoff is the transfer of a piece of work between people or agents.

In plain words
What is it for?
Use it when onboarding a new member or when a member has not yet completed the required introduction and first task.
Why use it?
It gives new members a consistent starting point and helps them begin useful work with another person.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: names the AskUserQuestion tool.

This is egregore-labs/egregore's own configuration. It tells Claude Code how to work on egregore itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything egregore configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is bash bin/person.sh onboard.

Reuse

Borrowing it

Nothing to install: this file belongs to egregore-labs/egregore. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/egregore-labs/egregore/main/.claude/skills/onboarding/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/egregore-labs/egregore

Made for: Claude Code.

Wrote 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.

agentmods badge for onboarding

README.md
[![agentmods](https://agentmods.dev/badge/skills/egregore-labs/egregore/onboarding/github.svg)](https://agentmods.dev/skills/egregore-labs/egregore/onboarding)
Your own site
<a href="https://agentmods.dev/skills/egregore-labs/egregore/onboarding"><img src="https://agentmods.dev/badge/skills/egregore-labs/egregore/onboarding/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.

agentmods 80×15 button for onboarding

Your own site · 80×15
<a href="https://agentmods.dev/skills/egregore-labs/egregore/onboarding"><img src="https://agentmods.dev/badge/skills/egregore-labs/egregore/onboarding.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 38 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,604 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 2 findings, up to high

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 →

  • high Privilege Escalation · line 112
    Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
    Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
  • high Privilege Escalation · line 113
    Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
    Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00038 $0.05604
Opus 5 $0.00019 $0.02802
Sonnet 5 $0.00008 $0.01121
Haiku 4.5 $0.00004 $0.00560

Measured 10d ago against content hash 2b26f0dbfa12, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

onboarding 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 10d 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.

- DO NOT call `curl` or any HTTP endpoint.
.claude/skills/onboarding/SKILL.md · 454 lines

How it starts

The opening of the file, as written. The whole thing — 454 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Welcome a new user to this Egregore.

Deterministic state machine: VERIFY → ORIENT → INVITE → FIRST_HANDOFF. Orient first, invite early — show what Egregore is, then show it works best with others. Profile data collected progressively during real work.

When to invoke

User says: "/onboarding", or is a new user whose onboarding_complete is false and needs to be verified, oriented, offered an invite, and walked through their first handoff.

Output discipline — CRITICAL

This is a conversation with a new user, not a CI pipeline.

Rules:

  • Batch all reads at once. Before each state, read ALL files you'll need in a single parallel tool call.
  • Batch all writes at once. State file update + API call = one parallel tool call group.
  • Suppress all command output. Every bash call must use 2>/dev/null or capture output.
  • No narration of internal steps. Never say "Let me read the config" or "Saving to state."
  • VERIFY should be invisible. All checks in one bash call. If everything passes, say nothing.

What the user should experience:

  1. A framing of what Egregore is (2-3 sentences from egregore.md)
  2. "It works best when there's someone on the other end."
  3. Name question (one AskUserQuestion)
  4. Creators only: Invite offer → invite someone or skip. Joiners skip this step entirely — they just got here; they shouldn't be bringing more people in before they've done any work.
  5. Save/handoff one-liner explanation
  6. "What are you working on?" → creates branch
  7. User works normally
  8. On session end: first handoff. Plus a gentle nudge about inviting others, creators only, if they skipped the invite offer.
  9. "You're in."

Local mode gate (applies to ALL states)

During VERIFY, check api_url from egregore.json. Store for entire flow.

If api_url is empty — local mode:

  • DO NOT call bin/graph.sh under any circumstances.
  • DO NOT call curl or any HTTP endpoint.
  • ONLY git operations are allowed.

If api_url is set — connected mode. All API calls proceed normally.

Read the full file on GitHub · 454 lines

Changes

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.

  1. 10d ago First seen · 454 lines · 38 tokens per session scan A 2b26f0dbfa12

Subscribe to this mod's changes

onboarding is a skill published in the GitHub repository egregore-labs/egregore (282 stars, last pushed 6d ago), licensed MIT. It adds 38 tokens to every session and 5,604 once invoked, about $0.0002 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 2026-08-30.