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 arrase/ollama-agent --skill task-creatorgit clone --depth 1 https://github.com/arrase/ollama-agentWrote 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/arrase/ollama-agent/task-creator)<a href="https://agentmods.dev/skills/arrase/ollama-agent/task-creator"><img src="https://agentmods.dev/badge/skills/arrase/ollama-agent/task-creator/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/arrase/ollama-agent/task-creator"><img src="https://agentmods.dev/badge/skills/arrase/ollama-agent/task-creator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00025 | $0.01114 |
| Opus 5.5 | $0.00010 | $0.00446 |
| Sonnet 5.5 | $0.00005 | $0.00223 |
| Haiku 4.5 | $0.00003 | $0.00111 |
Grade A, and why
task-creator 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 14d 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Task Creator
A saved task is a pre-configured, self-contained prompt template and configuration that can be executed autonomously on demand via /task run <task_id> [key=value ...] or ollama-agent task run <task_id> [key=value ...]. Tasks reside in /tasks/<task_id>.yaml (persisted in ~/.ollama-agent/tasks/<task_id>.yaml).
Task YAML Schema
Each task is saved as a single YAML file: /tasks/<task_id>.yaml.
title: "Clear, descriptive title of the task"
prompt: "Jinja2 prompt template with {{ variable }}, {% if %}, {% for %}, and @mentions..."
model: "gemma4:26b" # Required non-empty model name (e.g. the active model or a specific model)
reasoning_effort: "default" # Required reasoning effort (as advertised by the model in Ollama /api/show, e.g. "max", "low", "false", "hide"), or "default"
inputs:
target_file:
description: "Target file to analyze"
type: "string"
required: true
strict:
description: "Enable strict mode"
type: "boolean"
default: false
Fields Specification
title(string, required): Short descriptive title for display in/task list.prompt(string, required): Jinja2-compatible prompt template. Because tasks run autonomously, the prompt should provide all necessary context, step-by-step instructions, and expected deliverables.model(string, required): Non-empty Ollama model name (e.g."gemma4:26b"or the session's active model). Must not be empty ("").reasoning_effort(string, required): Non-empty reasoning effort supported by the target model in Ollama (advertised in/api/show), or"default"to use the model's default.inputs(mapping, optional): Dictionary of input parameter definitions:description(string, optional): Human-readable explanation of the input parameter.type(string, optional): Data type for value coercion (string,boolean,number). Default isstring.required(boolean, optional): Whether the variable must be provided when running the task. Default isfalse.default(any, optional): Default fallback value when the variable is omitted at runtime.
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.
- 14d ago Changed 6ad8d315cab3
- 1mo ago Changed adcbaf3c635e
- 2mo ago Changed · +57 lines d41c1e19e5e1
- 2mo ago First seen · 37 lines · 25 tokens per session scan A eddbe1234ad0
task-creator is a skill published in the GitHub repository arrase/ollama-agent (30 stars, last pushed 4d ago), licensed MIT. It adds 25 tokens to every session and 1,114 once invoked, about $0.0001 per session on Opus 5.5. 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.
task-creator compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| task-creatorarrase/ollama-agent | Skill | 30 | 4d ago | MIT | A | 25 | Claude Code, Codex |
| todosJetBrains/thinkrail | Skill | 7.9k | today | Apache-2.0 | A | 58 | Claude Code, Codex |
| computer-usekirodotdev/KiroCrew | Skill | 4.8k | today | Apache-2.0 | A | 62 | Claude Code, Codex |
| meetingskirodotdev/KiroCrew | Skill | 4.8k | today | Apache-2.0 | A | 62 | Claude Code, Codex |
| mochi-remindkirodotdev/KiroCrew | Skill | 4.8k | today | Apache-2.0 | A | 20 | Claude Code, Codex |
| agent-initializationPrism-Shadow/penguin-harness | Skill | 3.2k | today | Apache-2.0 | A | 79 | Claude Code, Codex |
| x-themex-cmd/x-cmd | Skill | 4.7k | 8d ago | Apache-2.0 | A | 74 | Claude Code, Codex |