setup-maintenance

A maintenance process for keeping a Pi coding-agent setup healthy over time. It checks for outdated versions, package conflicts, stale instructions, growing storage use, and possible improvements.

In plain words
What is it for?
Use it for monthly setup audits, before adding a package or extension, before publishing, or when checking whether the current agent setup is still healthy.
Why use it?
Agent setups can drift as packages, extensions, and project instructions change. Regular checks help reveal problems before they cause failures and help identify useful new capabilities.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/romiluz13/auto-pi/setup-maintenance
Any agent
npx skills add romiluz13/auto-pi --skill setup-maintenance
Clone the repo
git clone --depth 1 https://github.com/romiluz13/auto-pi

Made for: Claude Code, Codex.

Per session 62 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,222 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00062 $0.01222
Opus 5 $0.00031 $0.00611
Sonnet 5 $0.00012 $0.00244
Haiku 4.5 $0.00006 $0.00122

Measured 2d ago against content hash fef351417f45, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

setup-maintenance 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 2d 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.

skills/setup-maintenance/SKILL.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Setup Maintenance

A setup that doesn't degrade is a setup with a maintenance discipline. This skill defines the cadence + the two loops: maintenance (prevent drift) and improvement (find new leverage). The one-command runner is /setup-audit.

The two loops

Maintenance loop — prevent drift. Versions fall behind, new packages conflict, skills bloat, disk grows, AGENTS.md goes stale, extensions develop latent bugs. Without a check, you discover this when something breaks at the worst time.

Improvement loop — find new leverage. The Pi ecosystem ships new packages/features weekly. Other setups (cc10x, Archon, spences10, tomsej/pi-ext) invent techniques worth stealing. Without a scan, you plateau. This is the loop that turned a good setup into a best-contender.

The cadence

When What How
Monthly Full audit (both loops) /setup-audit — fans out 6 subagents, writes a report
On-add (new package/extension) Harmony re-audit Read the new package's tools/commands/hooks/storage; check for collisions against existing packages + extensions (enumerate dynamically from settings.json + extensions dir). Reject if it duplicates an existing axis.
Before-publish (pushing my-pi) Full audit + sync /setup-audit → fix criticals → sync settings/extensions/skills to ~/Dev/my-pi → commit + push
On-extend (new skill) Coach coverage check After adding a skill, grep coach.ts — does any intent route to it? If not, add a detector or it's unreachable via Coach.

The /setup-audit runner

/setup-audit dispatches 6 parallel read-only subagents:

  1. Version freshness — installed vs latest for Pi core + all npm packages. Flags drift + deprecation.
  2. Harmony re-audit — tool/command/hook/storage collisions across all packages + extensions. File:line evidence.
  3. Coach coverage scan — lists EVERY command + skill, flags which Coach never routes to. Each gap = leverage the setup has but doesn't surface. This is the highest-value axis — it's how we found the wayfinder gap.
  4. Disk growth — sessions, hermes DB, workflows, context sidecar. Flags unbounded stores > 100MB.
  5. AGENTS.md accuracy — line count (must be < 200) + does it match reality.
  6. Ecosystem scan — new Pi packages + public setups worth stealing from. Conflict-assessed.

Read the full file on GitHub · 71 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. 2d ago First seen · 71 lines · 62 tokens per session scan A fef351417f45

Subscribe to this mod's changes

setup-maintenance is a skill published in the GitHub repository romiluz13/auto-pi (11 stars, last pushed 17d ago), licensed MIT. It adds 62 tokens to every session and 1,222 once invoked, about $0.0003 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 2026-08-30.