What is thinking-toolkit AGENTS.md?

A set of instructions for Thinking Toolkit, a provider-neutral collection of practical thinking models for AI agents. It defines project rules, keeps the toolkit independent of hidden services, and preserves its model cards and categories.

Use it when maintaining, extending, testing, or documenting the Thinking Toolkit and its model cards, logic files, and localized readme files.

It prevents changes that would make the toolkit provider-specific, incomplete, or inconsistent across languages and files. It also keeps facts, assumptions, and hypotheses clearly separated.

thinking-toolkit AGENTS.md — Instructions file by ponomr

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/ponomr/thinking-toolkit/agents-md

Clone the repo

git clone --depth 1 https://github.com/ponomr/thinking-toolkit

Made for: Codex, OpenCode.

Per session 650 This file is loaded in full into every session.

When invoked 650 The same file — it is already loaded in full.

Security scan A 0 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.

Model	Per session	Once invoked
Fable 5	$0.00650	$0.00650
Opus 5	$0.00325	$0.00325
Sonnet 5	$0.00130	$0.00130
Haiku 4.5	$0.00065	$0.00065

Measured 2d ago against content hash 5875d07e0b10, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

thinking-toolkit AGENTS.md 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 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.

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.

AGENTS.md · 63 lines

How it starts

The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Thinking Toolkit — Standalone Context

A provider-neutral agent skill for selecting and applying practical thinking models. Works in any agent that supports the SKILL.md convention.

Critical Rules

Keep every project file in English, except the localized READMEs (README.ru.md, README.zh.md), which mirror README.md.
Keep the core skill usable by any capable LLM without provider-specific tools, browsing, code execution, persistent memory, or hidden state.
Never mention or cite the prohibited source project.
Never add external URLs or external Markdown links to the skill payload (SKILL.md, references/, logic/, agents/). READMEs and LICENSE are repository documentation and may cite external resources.
Preserve all 30 model cards and all four catalog categories.
The /logic module (logic/) is an English payload — no Cyrillic, even for trigger phrases; /logic recognizes and replies in the user's language by instruction, not by hardcoded examples. Keep it lean (4 files); it ships discipline, not a logic textbook.
Keep claims, assumptions, hypotheses, and user-provided facts distinct.
Do not commit changes unless the user explicitly requests a new commit.

Architecture

Path	Role
`SKILL.md`	Core adaptive workflow and direct reference routing
`references/catalog.md`	Complete model index, aliases, selection cues, and combinations
`references/*.md`	One detailed, progressively loaded card per model
`logic/*.md`	`/logic` argument-analysis module (overview + 3 references)
`scripts/validate_skill.py`	Deterministic structure and content validation
`tests/test_validate_skill.py`	Validator regression tests and project checks
`agents/openai.yaml`	Optional host-specific discovery metadata; not required by the core skill
`install.sh`	Copies the skill payload into a host agent's skills directory
`README.md`	Public repository documentation

Development Commands

Read the full file on GitHub · 63 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.

2d ago First seen · 63 lines · 650 tokens per session scan A 5875d07e0b10

Subscribe to this mod's changes

thinking-toolkit AGENTS.md is an instructions file published in the GitHub repository ponomr/thinking-toolkit (9 stars, last pushed 17d ago), licensed MIT. It adds 650 tokens to every session, about $0.0032 per session on Opus 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 2026-08-31.

Data and AI Localization Testing agent-skills claude-code codex decision-making mental-models openclaw

thinking-toolkit AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Thinking Toolkit — Standalone Context

Critical Rules

Architecture

Development Commands

What this file has done since we first saw it

Other instructions, from other repositories

skills CLAUDE.md

SkillCompass AGENTS.md

ClawFleet CLAUDE.md

cc-thinking-skills CLAUDE.md

autonomous-os AGENTS.md

GameDesignOS AGENTS.md