cron-script-fastpath

A code pattern for letting a scheduler run simple shell commands directly from a job definition instead of sending them through an AI model. It adds a script field and a scheduler branch for commands such as running a Python or Bash file.

In plain words
What is it for?
Use it when changing a scheduler so simple cron jobs can execute scripts directly while retaining their existing command output.
Why use it?
It removes an unnecessary AI step from straightforward scheduled jobs, reducing repeated model calls and avoiding related delay or timeout risk.

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/humanerd-drew/opencode-drewgent/cron-script-fastpath
Any agent
npx skills add humanerd-drew/opencode-drewgent --skill cron-script-fastpath
Clone the repo
git clone --depth 1 https://github.com/humanerd-drew/opencode-drewgent

Made for: Claude Code, Codex.

Per session 46 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,101 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 2 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 $0.00046 $0.05101
Opus 5 $0.00023 $0.02550
Sonnet 5 $0.00009 $0.01020
Haiku 4.5 $0.00005 $0.00510

Measured yesterday against content hash 290aa85c158b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cron-script-fastpath scanned grade A with 2 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s "${BASE_URL}/models" -H "Authorization: Bearer ${API_KEY}" \

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

directly via `subprocess.run()` and skips the LLM entirely.
skills/devops/cron-script-fastpath/SKILL.md · 468 lines

How it starts

The opening of the file, as written. The whole thing — 468 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Cron Script Fast-Path — LLM Bypass for Simple Shell Jobs

{{AGENT_NAME}} cron jobs run their prompt through AIAgent (LLM) by default. For jobs whose prompt is just a shell command ("Run: python3 xxx.py"), this is wasteful — LLM parses the prompt, runs the script, and forwards stdout to delivery. Each run costs a full LLM round-trip.

This skill documents a surgical patch that adds a script: field fast-path: when a job has script: "path", scheduler runs the script directly via subprocess.run() and skips the LLM entirely.


1. When to Use

Apply when a cron job's prompt is structurally a single shell command:

  • "Run: python3 ~/.{{AGENT_NAME_LOWER}}/scripts/foo.py"
  • "Run: bash ~/.{{AGENT_NAME_LOWER}}/scripts/bar.sh"
  • Any job where the prompt's only value-add over subprocess.run is "Report: ..." or "If silent, [SILENT]" — both of which the script already produces in stdout.

Two independent motivations:

Motivation Why When it bites hardest
Cost LLM round-trip per job tick — adds up on hourly/daily cadence with a 1M-context model. High-frequency jobs (every N minutes).
Reliability / timeout avoidance LLM-based cron jobs inherit the main model. If the provider is slow to start streaming or stalls mid-response, the cron scheduler's 600s idle timeout fires — job fails with TimeoutError: ... idle for 603s (limit 600s) — last activity: waiting for provider response (streaming). The script itself would finish in <30s, but the LLM wrapper around it times out on a slow API call. Any job where the default model can lag. The 10-min wall is tight — even a 4-min script + 5-min LLM stall = timeout.

Either motivation alone justifies converting. Reliability is often the stronger reason — a failing cron job is worse than a slightly more expensive one.

Do NOT use for jobs that need:

  • LLM reasoning (SEO article scoring, Trend analysis, site-audit synthesis)
  • Multi-step coordination with side tools (n8n, MCP, etc.)
  • Reading multiple files + writing a human-friendly report

Read the full file on GitHub · 468 lines

Files

What ships with it

3 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. yesterday First seen · 468 lines · 46 tokens per session scan A 290aa85c158b

Subscribe to this mod's changes

cron-script-fastpath is a skill published in the GitHub repository humanerd-drew/opencode-drewgent (2 stars, last pushed 1mo ago), licensed MIT. It adds 46 tokens to every session and 5,101 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 2 findings (makes network calls, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens