agentmods.dev

design-agent-tools

design-agent-tools is a skill for Claude Code, Codex from bikeread/promethos. It costs 26 tokens per session (699 once invoked), scanned A, original, MIT.

A guide for defining how an agent-facing tool should be used, what it accepts, what it returns, and how it fails. A tool is a callable action such as a file operation, API request, or database query.

In plain words
What is it for?
Use it when adding a tool, documenting one, or fixing unclear inputs, outputs, failure types, retry behavior, or side effects.
Why use it?
It reduces repeated misuse and makes errors clear enough for the agent to recover safely.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

not rated 33repo 7mo ago A scan 26 tokens original MIT

Good fit Use it when adding a tool, documenting one, or fixing unclear inputs, outputs, failure types, retry behavior, or side effects.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/bikeread/promethos/design-agent-tools
View source ↗ bikeread/promethos
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.

Any agent
npx skills add bikeread/promethos --skill design-agent-tools
Clone the repo
git clone --depth 1 https://github.com/bikeread/promethos

Made for: Claude Code, Codex.

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 design-agent-tools

README.md
[![agentmods](https://agentmods.dev/badge/skills/bikeread/promethos/design-agent-tools/github.svg)](https://agentmods.dev/skills/bikeread/promethos/design-agent-tools)
Your own site
<a href="https://agentmods.dev/skills/bikeread/promethos/design-agent-tools"><img src="https://agentmods.dev/badge/skills/bikeread/promethos/design-agent-tools/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 design-agent-tools

Your own site · 80×15
<a href="https://agentmods.dev/skills/bikeread/promethos/design-agent-tools"><img src="https://agentmods.dev/badge/skills/bikeread/promethos/design-agent-tools.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 26 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 699 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.
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 $0.00026 $0.00699
Opus 5.5 $0.00010 $0.00280
Sonnet 5.5 $0.00005 $0.00140
Haiku 4.5 $0.00003 $0.00070

Measured 2mo ago against content hash 458ab67a1511, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

design-agent-tools 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 2mo 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.

skills/design-agent-tools/SKILL.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Goal

Define a tool contract that is easy for an agent to use correctly and hard to misuse silently.

Inputs

  • Tool purpose
  • Upstream caller expectations
  • Expected failure conditions and side effects

Non-Goals

  • Implementing the tool runtime
  • Designing the entire agent around a single tool
  • Choosing the approval or autonomy policy that governs the tool

Workflow

Trigger signals

  • Agent calls a tool with wrong arguments repeatedly
  • Tool errors are vague and the agent cannot recover
  • User says "it keeps misusing this tool"
  • A new tool is being added without a documented contract

1. State the job the tool owns

Describe the single job the tool performs and the caller state it expects before invocation. Success criteria: The tool has one clear purpose and is not acting as a generic escape hatch.

2. Define input and output invariants

Specify what the tool accepts, what shape it returns, and what fields the caller may depend on every time. Success criteria: The contract is explicit enough to implement and consume without guessing field meanings.

3. Define error classes and retry behavior

Separate validation errors, environmental failures, permission denials, and transient faults. State what is safe to retry and what is not. Success criteria: The calling agent can distinguish recoverable from terminal failure paths.

4. Decide idempotency and replay behavior

State whether repeated calls with the same intent are safe, unsafe, or conditionally safe, and describe what the caller must do before retrying a mutating operation. Success criteria: Retry behavior and idempotency expectations are explicit, especially for writes and external side effects.

5. Make side effects and governing permission references visible

Document whether the tool reads, writes, mutates state, triggers external effects, and which existing permission or autonomy class the caller should apply when invoking it. Success criteria: The tool's risk profile is obvious from the contract without making the tool contract itself the owner of approval policy.

Read the full file on GitHub · 91 lines

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. 2mo ago First seen · 91 lines · 26 tokens per session scan A 458ab67a1511

Subscribe to this mod's changes

design-agent-tools is a skill published in the GitHub repository bikeread/promethos (33 stars, last pushed 7mo ago), licensed MIT. It adds 26 tokens to every session and 699 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.

Building agents
Related

design-agent-tools compared with skills from other repositories

design-agent-tools next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
design-agent-toolsbikeread/promethosSkill337mo agoMITA26Claude Code, Codex
imagegenopenai/codexSkill133kyesterdayApache-2.0A113Codex
playwright-climicrosoft/playwrightSkill99kyesterdayApache-2.0A19Claude Code
local-ai-agentsmicrosoft/ai-agents-for-beginnersSkill79k4d agoMITA200Claude Code, Codex
systematic-debuggingobra/superpowersSkill313k3d agoMITA21Claude Code
next-cache-components-adoptionvercel/next.jsSkill143ktodayMITA95Claude Code
next-partial-prefetching-adoptionvercel/next.jsSkill143ktodayMITA103Claude Code