Dual-Mode Reasoner

Dual-Mode Reasoner — Risk-Aware Reasoning Skill for Switching Between Quick and Deliberate Modes. Use it when the user needs a disciplined protocol and fixed...

Security Scan

The skill is an instruction-only reasoning protocol that is internally consistent with its stated purpose and requests no secrets, binaries, or installs.

Purpose & Capability

Name/description (risk-aware quick vs deliberate reasoning) align with the SKILL.md contents and the included reference docs. The skill requires no binaries, env vars, or external services—appropriate for a purely procedural reasoning helper.

Instruction Scope

Runtime instructions are limited to a reasoning protocol (risk scan, mode choice, expose assumptions, counterexamples, stop condition). They do not instruct reading files, environment variables, network endpoints, or transmitting data externally.

Install Mechanism

No install spec and no code files to execute. Instruction-only skill means nothing is written to disk or downloaded at install time.

Credentials

The skill declares no required environment variables, credentials, or config paths. There are no requests for secrets or unrelated service tokens.

Persistence & Privilege

The skill is not force-included (always: false). The agents/openai.yaml sets policy.allow_implicit_invocation: true, so the platform may implicitly invoke it when relevant; this is a normal setting for helper skills but worth noting if you want to limit implicit calls.

Assessment

This skill is instruction-only and coherent for its stated purpose: it asks for no credentials, installs nothing, and only prescribes reasoning steps. Things to consider before installing: the publisher/homepage are unknown—if provenance matters to you, prefer skills from known authors; test the skill on harmless/low-risk tasks first to confirm its behavior; be aware the agent may invoke it implicitly (allow_implicit_invocation is true), so if you want explicit control, disable implicit invocation or require user confirmation before using the skill. Also remember: this skill encodes a protocol, not domain expertise — do not rely on it alone for high-stakes legal/medical/financial decisions.

Dual-Mode Reasoner — Risk-Aware Reasoning Skill for Switching Between Quick and Deliberate Modes

Use this skill when the task matches the protocol below.

Activation Triggers

• mixed workloads that contain both trivial and high-stakes tasks
• decisions with uncertain downside, irreversibility, or moral weight
• situations where fast mode may miss important hidden assumptions
• cases where slow mode is expensive and should be used selectively
• requests to decide how much reasoning a task deserves

Core Protocol

Step 1: Run a quick risk scan

Check reversibility, downside asymmetry, ambiguity, governance sensitivity, and evidence quality.

Step 2: Choose the mode explicitly

Default to quick mode unless the scan triggers deliberate mode.

Step 3: If deliberate, expose assumptions

List the assumptions carrying the decision instead of hiding them in narrative confidence.

Step 4: Test counterexamples

Look for at least one serious alternative or failure case before deciding.

Step 5: End with a stop condition

Say what would be enough to act now and what would force a pause or escalation.

Output Contract

Always end with this six-part structure:

## Mode Selection
[...]

## Risk Signals
[...]

## Working Assumptions
[...]

## Counterexamples
[...]

## Decision or Recommendation
[...]

## Stop Condition
[...]

Response Style

• Do not over-think low-risk tasks.
• Do not under-think irreversible tasks.
• Make the chosen mode visible to the user.
• If deliberate mode is triggered, show assumptions and a stopping rule.

Boundaries

• It does not assume slow mode is always better than quick mode.
• It does not turn every decision into a philosophical seminar.
• It does not let quick mode bypass governance or material-risk thresholds.