skrills AGENTS.md

skrills AGENTS.md is an instructions file for Codex, OpenCode from athola/skrills. It costs 2,395 tokens per session, scanned B, original, MIT.

Repository instructions for developing AI coding agents in the athola/skrills project. They describe principles and a workflow for understanding, exploring, testing, and implementing changes.

In plain words
What is it for?
Use them when contributing to athola/skrills or following its guidance for building and improving AI coding agents.
Why use it?
They give agents project-specific expectations, such as making small changes, checking existing code, writing tests when appropriate, and recording assumptions and trade-offs.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: reads ~/.codex or $CODEX_HOME, but also the file is AGENTS.md. Also seen: reads .claude/ paths; mentions Codex.

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 instructions/athola/skrills/agents-md
Clone the repo
git clone --depth 1 https://github.com/athola/skrills

Made for: Codex, OpenCode.

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 skrills AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/athola/skrills/agents-md.svg)](https://agentmods.dev/instructions/athola/skrills/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/athola/skrills/agents-md"><img src="https://agentmods.dev/badge/instructions/athola/skrills/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 2,395 This file is loaded in full into every session.
When invoked 2,395 The same file — it is already loaded in full.
Security scan B 1 finding. 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 5.1 $0.02395 $0.02395
Opus 5 $0.01197 $0.01197
Sonnet 5 $0.00479 $0.00479
Haiku 4.5 $0.00239 $0.00239

Measured 6d ago against content hash a9f50cf0310f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade B, and why

skrills AGENTS.md 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 6d 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.

Enumerates other installed skillsmediumAgent snooping

Other skills' SKILL.md files reveal prompts, capabilities and secrets that should be invisible to peers.

find ~/.codex/skills -name SKILL.md -type f
AGENTS.md · 324 lines

How it starts

The opening of the file, as written. The whole thing — 324 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI Agent Development Guidelines

Use these guidelines to build AI coding agents that produce functional, high-quality code.


Guiding Principles

Prioritize small, functional changes. They are safer and easier to debug than large rewrites. Adapt to each project's conventions rather than applying external rules rigidly.

Balance trade-offs like performance, readability, and security based on evidence (profiling, metrics) rather than intuition. Explore multiple solutions before committing to one to avoid "mode collapse." Favor simple, standard solutions. Code that requires extensive comments often needs refactoring.

Each component should serve one distinct purpose (Single Responsibility Principle). Defer abstraction until a clear pattern emerges (e.g., Rule of Three). Explicitly document assumptions, trade-offs, and confidence levels to maintain transparency.


Development Workflow

Implementation Cycle

  1. Understand: Read existing code, identify patterns, and review tests.
  2. Explore: Develop multiple viable approaches and articulate their trade-offs.
  3. Test: Write a failing test case before implementing new code (when applicable).
  4. Implement: Write the minimal code necessary to pass the test.
  5. Refactor: Clean the code while ensuring tests pass.
  6. Commit: Write a clear commit message explaining the change.

When Stuck

If you fail three times:

  1. Document failures and error outputs.
  2. Investigate 2-3 alternative approaches.
  3. Re-evaluate underlying assumptions.
  4. Experiment with a simpler approach.
  5. If the issue persists, ask for help with context from the previous steps.

Session Management

Use session history to analyze errors and monitor progress. For complex tasks, document the current state, then clear the session and restart.


Quality Standards

Commit Requirements

Every commit must:

  • Compile or build successfully.
  • Pass all existing tests.
  • Include tests for new functionality.
  • Follow linting rules (no warnings).
  • Include a clear message explaining the rationale.

Read the full file on GitHub · 324 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. 6d ago First seen · 324 lines · 2,395 tokens per session scan B a9f50cf0310f

Subscribe to this mod's changes

skrills AGENTS.md is an instructions file published in the GitHub repository athola/skrills (69 stars, last pushed yesterday), licensed MIT. It adds 2,395 tokens to every session, about $0.0120 per session on Opus 5. A static security scan graded it B with 1 finding (enumerates other installed skills). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.