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 uchicago-dsi/ai-sci-skills --skill maintenance-passgit clone --depth 1 https://github.com/uchicago-dsi/ai-sci-skillsWrote 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/uchicago-dsi/ai-sci-skills/maintenance-pass)<a href="https://agentmods.dev/skills/uchicago-dsi/ai-sci-skills/maintenance-pass"><img src="https://agentmods.dev/badge/skills/uchicago-dsi/ai-sci-skills/maintenance-pass/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/uchicago-dsi/ai-sci-skills/maintenance-pass"><img src="https://agentmods.dev/badge/skills/uchicago-dsi/ai-sci-skills/maintenance-pass.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 Excessive Agency · line 119 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00083 | $0.03009 |
| Opus 5 | $0.00042 | $0.01504 |
| Sonnet 5 | $0.00017 | $0.00602 |
| Haiku 4.5 | $0.00008 | $0.00301 |
Grade A, and why
maintenance-pass 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 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.
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 — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Maintenance Pass
Optimize For Maintenance Yield
Prioritize the safest change with the largest maintenance payoff. Delete or simplify code only when evidence supports it. Refactor only when it materially reduces duplication, clarifies ownership, removes a failed path, or protects an important contract. Do not move code around for aesthetics.
Use this rough ranking:
maintenance yield = impact * confidence * stale_touch_signal / blast_radius
Treat time since last git touch as a triage signal, not proof. The relevant metric is the most recent commit that changed a path, not when the file was created. Code whose latest git touch was a long time ago is more likely to be a vestige, duplicated old path, or stale interface, but stable core infrastructure can also be old and correct.
Start With Evidence
- Read local repo instructions first:
AGENTS.md,CLAUDE.md,README.md, nearby runbooks, and provenance notes tied to the target area. - Check the worktree:
git status --short --untracked-files=all. Never revert unrelated user changes. Avoid destructive cleanup in a dirty worktree unless the user explicitly scopes it. - Establish stable baselines:
git ls-files '*.py' | wc -l
git ls-files '*.py' | xargs wc -l | tail -n 1
- Build a least-recently-touched candidate list from tracked files before choosing a theme. Prefer files and directories whose most recent git commit touch is old, especially when they are scripts, one-off analyses, retired mechanism code, stale configs, duplicate helpers, or docs for old paths. Use
git log -1 --format='%ct %cs %h' -- <path>or the inventory helper's least-recently-touched section. Do not use last-touch age alone to delete stable package owners, tests, schemas, or public APIs. - If the user gives a target LOC change, treat it as a stopping target, not a quota. Aim for the target with the safest high-yield changes first, but stop short when the next candidate has weak evidence or excessive blast radius. Do not pad the diff with formatting, moves, or speculative rewrites to hit the number.
- Establish a pre-pass commit boundary before editing without creating empty commits. If the worktree is clean, record the current
HEADas the starting boundary instead of making an empty checkpoint. If the worktree is dirty, do not silently commit unrelated or user-owned changes; either commit only already-approved scoped changes as a real pre-pass checkpoint, or record the currentHEADplus the dirty status as the boundary. If the user explicitly asks for before/after commits but there is nothing real to checkpoint before the pass, report that the starting boundary is the existingHEAD; do not make an empty commit. - Optionally run the bundled inventory helper from the repo root:
What ships with it
2 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.
- 10d ago First seen · 188 lines · 83 tokens per session scan A ac6d02cc264b
maintenance-pass is a skill published in the GitHub repository uchicago-dsi/ai-sci-skills (17 stars, last pushed yesterday), licensed MIT. It adds 83 tokens to every session and 3,009 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
diagnose
Trace from a reproduced symptom to the source code that causes it. Pin the specific file and approximate line, rate confidence in the cause and clarity of the fix independently, and always propose a concrete fix.
repro-admin
Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
byted-util-volcengine-detect-retry
An orchestration workflow for Volcengine Cloud Detect, a service that checks websites or network endpoints from test locations.