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 agentmods add skills/wrg32786/aigent-os/setupnpx skills add wrg32786/aigent-os --skill setupgit clone --depth 1 https://github.com/wrg32786/aigent-osWhat 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 | $0.00023 | $0.00679 |
| Opus 5 | $0.00012 | $0.00340 |
| Sonnet 5 | $0.00005 | $0.00136 |
| Haiku 4.5 | $0.00002 | $0.00068 |
Grade A, and why
setup 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 yesterday.
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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
aigent-OS setup and reconfiguration
/start owns the day-one onboarding arc. /setup is the deeper configuration flow and may also be used later to revise one section.
Entry behavior
Read .aigent/state.json first.
- If state is missing or not
ready, route through/startunless the operator explicitly requested advanced setup. - If state is
ready, ask which configuration area needs revision rather than replaying the entire interview. - Never detect setup state by searching for placeholder prose in a Markdown file.
Before writing, set status: setup-in-progress while preserving any prior completedAt. On interruption, the next /start or /setup should resume from the last durably completed section.
Interview
Ask one question at a time and write each completed section immediately.
1. Identity
Ask for the operator's role, responsibilities, preferred collaboration style, and risk posture. Update system/00_identity.md without replacing framework-owned doctrine.
2. Priorities
Ask for the top two or three active priorities, the most urgent outcome, operating mode, and blockers. Update vault/memory/ACTIVE_PRIORITIES.md.
3. Authority boundaries
Ask what the AI may handle autonomously, what needs confirmation, what is human-only, and whether spending has an escalation threshold. Update system/12_authority_matrix.md.
4. Decision logic
Ask what makes the operator accept or reject opportunities, their active-work limit, and their most common time trap. Update system/14_decision_framework.md.
5. Projects and people
For each active project, capture purpose, current state, priority, next action, and relevant people. Create or update notes under vault/projects/ and vault/people/, then connect them from active priorities with wikilinks.
6. Specialist agents
Ask whether any specialized agents are useful now. For each accepted agent, capture name, scope, tools, model tier, escalation boundary, and success criteria. Store the definition in vault/agents/ with valid name: and tools: frontmatter so the installer can register it in .claude/agents/.
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.
- yesterday First seen · 78 lines · 23 tokens per session scan A f830e8cddaea
setup is a skill published in the GitHub repository wrg32786/aigent-os (16 stars, last pushed 7d ago), licensed MIT. It adds 23 tokens to every session and 679 once invoked, about $0.0001 per session on Opus 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 2026-09-01.
Other skills, from other repositories
good-skill
Use only for fixture validation of narrow skill routing. Avoid for normal repo work.
broad-skill
Always use this skill for every task and all tasks.
caveman-fixture
Always use caveman terse speech for every task.
missing-description
No description.
deep-domain-learning
Use when building deep understanding of a new domain, producing a world model document, preparing for expert-level conversations, or needing to reason from first principles in an unfamiliar field. Encodes the Why-Chain Mandate, Quantification Mandate, Evidence Discipline, Grounding Requirement, and Narrative Flow…
chat-selfie
Give your AI Agent a face and a heart. Use AI image generation or mood-mapped local sticker assets to let the agent proactively send emotional selfies that visualize its feelings during conversation.