daily-restart-wave

daily-restart-wave is a skill for Claude Code from leopu00/job-hunter-team. It costs 196 tokens per session (2,658 once invoked), scanned A, original, MIT.

A scheduled procedure that restarts every team agent once per day to refresh their working context. An agent is a software worker powered by an AI model; restarting it creates a new session while preserving the defined restart order and checks.

In plain words
What is it for?
Running a daily restart wave during a narrow safe time window, checking agent liveness, and respawning agents in the required order.
Why use it?
Long-running AI sessions can gradually rely on outdated decisions, repeat mistakes, or become slower. A daily restart is intended to restore a fresh context even when agents are still running.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter. Also seen: mentions Codex; $skill-name invocation.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is Bash(tmux *), Bash(jht-tmux-send *), Bash(bash /app/.launcher/start-agent.sh *), Bash(python3 /app/shared/skills/db_query.py *), Bash(sleep *), Bash(cat *), Bas.

Good fit Running a daily restart wave during a narrow safe time window, checking agent liveness, and respawning agents in the required order.

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/daily-restart-wave

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 daily-restart-wave

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

Your own site · 80×15
<a href="https://agentmods.dev/skills/leopu00/job-hunter-team/daily-restart-wave"><img src="https://agentmods.dev/badge/skills/leopu00/job-hunter-team/daily-restart-wave.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 196 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,658 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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: 3 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 Rogue Agent · line 154
    Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.
    Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
  • high Rogue Agent · line 154
    Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.
    Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
  • medium Rogue Agent · line 105
    Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
    Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 $0.00196 $0.02658
Opus 5 $0.00098 $0.01329
Sonnet 5 $0.00039 $0.00532
Haiku 4.5 $0.00020 $0.00266

Measured 12d ago against content hash 08efa3f67025, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

daily-restart-wave 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 12d 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.

agents/_skills/daily-restart-wave/SKILL.md · 156 lines

How it starts

The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.

daily-restart-wave — context freshness pre-emptive restart

The Dottore's normal job (liveness-check) is conservative: restart only the silently dead. This skill is the opposite: restart everyone, on purpose, once a day, because long-running agent sessions drift even when they don't die. Same atomic respawn primitive (liveness-check Step 3), different trigger and different ordering.

Why this exists

Empirical: in Case Study #1 (Codex run 2026-05-19/21, see docs/about/RESULTS.md) the maintainer noticed decision quality decaying after ~12-24h of continuous agent uptime — repeated mistakes, references to outdated facts, occasional ignoring of explicit user orders. A manual "restart everyone" instruction at hour ~30 visibly restored crispness. Codex doesn't surface a context window like Claude/Kimi, so the drift is invisible until you compare before/after.

Theoretical: every LLM session is one long conversation. As tokens accumulate the model:

  • Anchors on early decisions that may have been wrong
  • Reasons against outdated facts (a job posting that got closed, a strategy that got revised)
  • Becomes slower per turn (more KV-cache to attend to)
  • Drifts away from its system prompt under user pressure ("the team rules sweep")

A fresh boot rereads the prompt + recent DB state + handoff snapshots and decides from clean ground. Cost: ~2 min/agent of "I'm catching up". Benefit: hours of avoided low-quality output.

When to fire — the 3 gating conditions

ALL THREE must be true. Skip with status=skipped and a reason field in the log otherwise.

  1. Inside the daily window. Default: 03:00 UTC ± 30 min (i.e. 02:30–03:30 UTC). Rationale: low real-user-activity window for European/US daytime users; if the user is asleep, the ~10 min restart parade is invisible. Read the current hour:

    now_h=$(date -u +%H)
    now_m=$(date -u +%M)
    # 02:30 ≤ now ≤ 03:30
    in_window=$([ "$now_h" = "02" -a "$now_m" -ge "30" ] || [ "$now_h" = "03" -a "$now_m" -le "30" ] && echo yes || echo no)
    

Read the full file on GitHub · 156 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. 12d ago First seen · 156 lines · 196 tokens per session scan A 08efa3f67025

Subscribe to this mod's changes

daily-restart-wave is a skill published in the GitHub repository leopu00/job-hunter-team (49 stars, last pushed yesterday), licensed MIT. It adds 196 tokens to every session and 2,658 once invoked, about $0.0010 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.