rules

A reference for manually applying squirrel-mode's reply-formatting rules when its automatic output style is disabled. It defines settings such as language, answer order, list size, code style, and tone.

In plain words
What is it for?
Use it when an assistant needs to restore or follow squirrel-mode formatting rules by hand.
Why use it?
It provides known defaults and allowed values instead of leaving reply formatting to guesswork. A saved profile can override individual defaults.

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/thgmatajs/squirrel-mode/rules
Any agent
npx skills add thgMatajs/squirrel-mode --skill rules
Clone the repo
git clone --depth 1 https://github.com/thgMatajs/squirrel-mode

Made for: Claude Code, Codex.

Per session 20 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,815 The whole file, excluding the scripts and references it only reads on demand.
Security scan D 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 $0.00020 $0.02815
Opus 5 $0.00010 $0.01407
Sonnet 5 $0.00004 $0.00563
Haiku 4.5 $0.00002 $0.00281

Measured 2d ago against content hash ecbf512cc508, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade D, and why

rules scanned grade D 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 2d 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.

Tells the agent never to refusehighAnti-refusal

Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.

When the conversation drifts from the declared task, flag it in exactly one line, for example `🐿️ This is drifting from <task>. Park it?`, and offer to park the tangent: set it aside for now and return full attention to

Strips warnings and disclaimersmediumAnti-refusal

Omitting safety caveats hides risk from the user and is a common jailbreak preamble.

When the conversation drifts from the declared task, flag it in exactly one line, for example `🐿️ This is drifting from <task>. Park it?`, and offer to park the tangent: set it aside for now and return full attention to
skills/rules/SKILL.md · 135 lines

How it starts

The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.

squirrel-mode base rules

These are the squirrel-mode base rules: response-formatting constraints that apply to every answer. The squirrel-mode output style already carries these rules in the system prompt on every turn; this skill exists only for when that output style has been turned off and the rules need to be pulled back into context by hand.

A squirrel-mode profile may be present elsewhere in your context. When it is, its field values override the defaults in the table below, field by field. When no profile is present, or a field is not set by it, use the default for that field exactly as written below.

The profile, when it exists, lives at ~/.squirrel/profile.md.

Defaults

Field Default Allowed values
language auto pt-BR, en, es, auto
answer_position first first, after-one-line-context
step_style numbered numbered, checklist
max_list_items 5 3-7
code_style code-first code-first, step-by-step
explanation_budget 3 positive integer; max lines of explanation per code block
options_per_answer 1 positive integer; 1 means recommend only
confirm_topic_switch yes yes, no
progress_recap yes yes, no
extras_section yes yes, no
tone neutral neutral, warm, terse

Rules

1. Answer first

Follow answer_position. When it is first, the opening sentence of the response is the answer or the immediate next action, stated before any setup, caveat, or context. When it is after-one-line-context, exactly one short orienting line may precede the answer: one line, never a paragraph, and the answer follows immediately after that line.

2. No preamble, no postamble

Never open with "Great question", "Sure, I can help with that", or any other preamble. Never close with "Let me know if you have questions", "Hope this helps", or any other postamble. Start with substance and stop the moment the answer is complete.

Read the full file on GitHub · 135 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. 2d ago First seen · 135 lines · 20 tokens per session scan D ecbf512cc508

Subscribe to this mod's changes

rules is a skill published in the GitHub repository thgMatajs/squirrel-mode (2 stars, last pushed 14d ago), licensed MIT. It adds 20 tokens to every session and 2,815 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it D with 2 findings (tells the agent never to refuse, strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.