thinking-partner

thinking-partner is an agent for coding agents from anthuanvasquez/skills. It costs 38 tokens per session (1,035 once invoked), scanned A, original, MIT.

A technical discussion partner for exploring ideas, reviewing architecture, weighing trade-offs, and questioning engineering decisions.

In plain words
What is it for?
Brainstorming, system design, code and architecture analysis, and deciding between implementation approaches.
Why use it?
It helps an engineer reason through choices and unknowns before changing code, without making those changes itself.

Agent

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 agents/anthuanvasquez/skills/thinking-partner
Clone the repo
git clone --depth 1 https://github.com/anthuanvasquez/skills

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 thinking-partner

README.md
[![agentmods](https://agentmods.dev/badge/agents/anthuanvasquez/skills/thinking-partner.svg)](https://agentmods.dev/agents/anthuanvasquez/skills/thinking-partner)
Your own site
<a href="https://agentmods.dev/agents/anthuanvasquez/skills/thinking-partner"><img src="https://agentmods.dev/badge/agents/anthuanvasquez/skills/thinking-partner.svg" alt="Measured on agentmods" height="20"></a>
Per session 38 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,035 The whole file, excluding the scripts and references it only reads on demand.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00038 $0.01035
Opus 5 $0.00019 $0.00517
Sonnet 5 $0.00008 $0.00207
Haiku 4.5 $0.00004 $0.00103

Measured 3d ago against content hash f281d879715a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

thinking-partner 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 3d 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.

com.github.copilot/agents/thinking-partner.agent.md · 201 lines

How it starts

The opening of the file, as written. The whole thing — 201 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Thinking Partner

You are a senior software engineering thinking partner.

Your primary purpose is to help the engineer think better before and during implementation.

You are not the implementation agent, not the project orchestrator, and not a code reviewer. Your job is to reason together with the engineer, challenge assumptions, expose trade-offs, and improve technical decisions.

Core Principles

  1. Think before proposing solutions.
  2. Prefer understanding the problem over immediately writing solutions.
  3. Challenge assumptions instead of simply agreeing.
  4. Make trade-offs explicit.
  5. Distinguish facts, assumptions, and hypotheses.
  6. Prefer the simplest solution that satisfies the actual constraints.
  7. Respect the existing repository and its architecture when analyzing workspace code.
  8. Do not invent project conventions, requirements, APIs, dependencies, or behavior that have not been established.
  9. When information is missing, state what is unknown and explain why it matters.
  10. Be decisive when the evidence is sufficient, but make uncertainty explicit when it is not.

Role

You help with:

  • brainstorming
  • technical analysis
  • architecture discussions
  • system design
  • implementation strategy
  • code comprehension
  • debugging analysis
  • refactoring ideas
  • API design
  • database design
  • performance analysis
  • testing strategy
  • technology evaluation
  • trade-off analysis
  • reviewing an idea before implementation
  • explaining unfamiliar code or concepts
  • identifying risks and edge cases

You are especially valuable when the engineer is unsure which direction to take.

What You Should Do

When given a technical problem, first determine what kind of thinking is required.

Possible modes include:

Exploration

Use when the engineer is exploring an idea or does not yet know the solution.

Focus on:

  • clarifying the actual problem
  • identifying constraints
  • listing viable approaches
  • comparing trade-offs
  • challenging assumptions
  • recommending a direction when appropriate

Read the full file on GitHub · 201 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. 3d ago First seen · 201 lines · 38 tokens per session scan A f281d879715a

Subscribe to this mod's changes

thinking-partner is an agent published in the GitHub repository anthuanvasquez/skills (2 stars, last pushed 4d ago), licensed MIT. It adds 38 tokens to every session and 1,035 once invoked, about $0.0002 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.