gpd-settings

Configure GPD workflow toggles, tier model overrides, review cadence, and git branching preferences.

Updated Mar 18, 2026
One-click install
npx skills add https://github.com/MichaelsEngineering/sentinel-multiscale-field-dynamics --skill gpd-settings-michaelsengineering
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gpd-settings
Source: https://github.com/MichaelsEngineering/sentinel-multiscale-field-dynamics/tree/main/.agents/skills/gpd-settings
Command: npx skills add https://github.com/MichaelsEngineering/sentinel-multiscale-field-dynamics --skill gpd-settings-michaelsengineering

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Managing GPD workflow configuration manually requires editing .gpd/config.json by hand and knowing every valid key, which is error-prone and easy to get wrong across different runtimes. ## Core Features & Use Cases - Interactive Settings Prompt: Presents a single multi-question form covering autonomy level, research mode, research profile, workflow agent toggles, review cadence, parallelization, and git branching strategy. - Runtime-Aware Tier Model Overrides: Detects the active runtime and lets you pin explicit tier-1, tier-2, and tier-3 model strings in that runtime's native identifier format, or fall back to runtime defaults. - Config Merging and Confirmation: Reads existing .gpd/config.json, merges new answers without clobbering other runtimes' overrides, and displays a summary table with quick command references. - Use Case: After installing GPD in a new repository, run the settings workflow once to choose a balanced autonomy level, enable the plan checker and verifier, and set per-phase git branching before starting plan-execute cycles. ## Quick Start Ask the AI to run the GPD settings workflow to configure research profile, tier models, and workflow toggles for this project.

Frequently Asked Questions about gpd-settings

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I configure GPD workflow settings for my project?▼

Run the gpd-settings workflow, which presents one interactive prompt covering autonomy, research mode, profile, tier models, workflow toggles, cadence, parallelization, and git branching. Answers are merged into .gpd/config.json and confirmed with a summary table.

How do I set explicit tier-1, tier-2, and tier-3 model overrides in GPD?▼

Choose 'Configure explicit tier models' in the Tier Models question, then provide the exact model strings your active runtime accepts. The skill preserves provider prefixes, slashes, brackets, and alias syntax without normalization beyond trimming whitespace.

Does GPD settings store project conventions like units or notation?▼

No. Project conventions live in .gpd/CONVENTIONS.md and .gpd/state.json via the convention_lock, managed with gpd convention set. The settings workflow explicitly avoids writing a physics section into .gpd/config.json.

What happens if the active runtime cannot be detected for model overrides?▼

The workflow infers the runtime from command syntax, tool names, environment, and local config directories. If it remains ambiguous, it asks you which runtime to configure before continuing with model override questions.

What review cadence options does GPD execution support?▼

Three options: dense for frequent bounded review points, adaptive (default) for automatic first-result and risky-fanout gates, and sparse for the fewest stops while required correctness gates still run. Cadence is independent of model profile and research mode.