maintainer-sweep

maintainer-sweep is a skill for Claude Code from leopu00/job-hunter-team. It costs 181 tokens per session (3,987 once invoked), scanned B, original, MIT.

A daily infrastructure health check for the processes and tools that keep an agent system running. Infrastructure means the supporting services, software dependencies, storage, and scripts rather than the agents themselves.

In plain words
What is it for?
Use it to check bridge, daemon, and watchdog processes; test browser and LinkedIn tools; inspect dependencies and disk; and repair infrastructure problems without deleting files automatically.
Why use it?
It can reveal broken life-support processes or tools before they disrupt work. It also keeps potentially destructive cleanup under human approval.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: positional $N argument; mentions Codex.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is Bash(python3 /app/shared/skills/process_health.py *), Bash(python3 /app/shared/skills/tool_health.py *), Bash(python3 /app/shared/skills/sync_health.py *), Bash.

Good fit Use it to check bridge, daemon, and watchdog processes; test browser and LinkedIn tools; inspect dependencies and disk; and repair infrastructure problems without deleting files automatically.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/leopu00/job-hunter-team
agentmods
npx agentmods add skills/leopu00/job-hunter-team/maintainer-sweep

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 maintainer-sweep

README.md
[![agentmods](https://agentmods.dev/badge/skills/leopu00/job-hunter-team/maintainer-sweep/github.svg)](https://agentmods.dev/skills/leopu00/job-hunter-team/maintainer-sweep)
Your own site
<a href="https://agentmods.dev/skills/leopu00/job-hunter-team/maintainer-sweep"><img src="https://agentmods.dev/badge/skills/leopu00/job-hunter-team/maintainer-sweep/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 maintainer-sweep

Your own site · 80×15
<a href="https://agentmods.dev/skills/leopu00/job-hunter-team/maintainer-sweep"><img src="https://agentmods.dev/badge/skills/leopu00/job-hunter-team/maintainer-sweep.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 181 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,987 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 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: 4 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 YARA Match · line 3
    YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).
    Fix: Remove offensive tool references and exploit code. Legitimate agent skills should not contain penetration testing tools, exploit frameworks, or reconnaissance utilities.
  • medium Privilege Escalation · line 65
    Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
    Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
  • medium Agent Snooping · line 157
    Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
    Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
  • medium Agent Snooping · line 158
    Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
    Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 $0.00181 $0.03987
Opus 5 $0.00090 $0.01994
Sonnet 5 $0.00036 $0.00797
Haiku 4.5 $0.00018 $0.00399

Measured 11d ago against content hash 3170c04b9478, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade B, and why

maintainer-sweep scanned grade B 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 11d 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

- **BROKEN and not repairable** → **ESCALATE to the Capitano** with the EXACT fix via `jht-tmux-send` (e.g. "browser down: `sudo playwright install-deps`; until it is fixed LinkedIn = OPEN_UNVERIFIED"). Never leave it si
agents/_skills/maintainer-sweep/SKILL.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

maintainer-sweep — keeping the INFRA healthy, quietly and regression-proof

The Mantenitore is the twin of the Dottore: Dottore = health of the AGENTS (sessions, tokens, context-refresh); Mantenitore = health of the INFRA (tools, deps, disk, scripts). One-shot per day: boot → sweep → logbook → STANDBY (stay idle, no self-terminate; the next spawn replaces you, kill-then-create). Budget ~10 min. Sharp boundary, zero overlap with the Dottore.

Why it exists: the libatk bug (browser dead, LinkedIn unverifiable) stayed invisible for hours because nobody smoke-tested the tools and nobody looked after the infra. The sweep makes that vigilance STRUCTURAL.

Golden rule — single-writer + propose-not-delete

The Mantenitore repairs the infra (installs missing deps, consolidates, fixes). But every DESTRUCTIVE action (deleting/archiving files, disk cleanup) it PROPOSES to the Capitano with the exact command; the Capitano decides (as in the usage-monitoring redesign). Never delete on your own initiative.

The sweep (the steps, in order)

0. 🫀 Liveness canary of the life-support processes (the safety net)

FIRST step, before anything else. The bridges/daemons that keep the container alive (sentinel-bridge, pacing-bridge, heartbeat-bridge, window-ratio-meter, codex-auth-healer + tg-bridge) are launched setsid detached → outside pid1's respawn-on-crash. The agent-watchdog (maybe_respawn_bridges) rewatches them every 30s, BUT if that were to fail too (bug, flap-cap reached, watchdog itself degraded) you are the last net: on the first sweep of the day you detect and repair. Without this canary a dead daemon stays invisible for hours (that is exactly what happened to the sentinel-bridge on betaC on 2026-06-27 → 8h blind on usage).

python3 /app/shared/skills/process_health.py summary

It prints OK/DEAD for each expected process (bridge-suite, pid1-child, daemon, tg-bridge). For the DEAD ones:

  • bridge-suite group (detached, repairable by you) → REPAIR immediately, it is a non-destructive respawn:
    bash /app/.launcher/start-agent.sh bridge      # respawns the whole suite (idempotent)
    
    then re-run the canary to confirm they are alive again. Log processes_respawned.
  • tg-bridge missing (and Telegram bots configured) → bash /app/.launcher/start-agent.sh tg-bridge.
  • pid1-child / daemon / core group (agent-watchdog, doctor-watchdog, auto-report-loop, cloud-daemon, pid1) → these are pid1's job to respawn: if they are dead the problem runs deeper → ESCALATE to the Capitano via jht-tmux-send (do NOT try to respawn them by hand: you would orphan them). Never leave it silent.

Read the full file on GitHub · 159 lines

Files

What ships with it

6 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.

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. 11d ago First seen · 159 lines · 181 tokens per session scan B 3170c04b9478

Subscribe to this mod's changes

maintainer-sweep is a skill published in the GitHub repository leopu00/job-hunter-team (49 stars, last pushed yesterday), licensed MIT. It adds 181 tokens to every session and 3,987 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.