skills.rest
rmethodm
rmethodmCOMMUNITY
Explore all skills by this author →
GitHub

ss-resolve

Compile a deterministic StyleSeed rule bundle for one agent from a design lock file.

Updated May 26, 2026
Search Tags:#design tokens#design rules#agent context#styleseed#context compiler#node script#ui grammar
One-click install
npx skills add https://github.com/rmethodm/resumegen --skill ss-resolve-rmethodm
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ss-resolve
Source: https://github.com/rmethodm/resumegen/tree/main/.agents/skills/ss-resolve
Command: npx skills add https://github.com/rmethodm/resumegen --skill ss-resolve-rmethodm

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Loading a full design rule handbook into every agent turn wastes context and produces inconsistent results. This Skill compiles a small, deterministic StyleSeed rule bundle tailored to one agent, output grammar, surface adapter, domain, page type, brand recipe, palette recipe, and optional profile. ## Core Features & Use Cases - Deterministic context compilation: Runs scripts/resolve-context.mjs to emit .styleseed/effective-rules.md plus a manifest.json recording selections, source hashes, and the bundle hash. - Lock-driven or flag-driven resolution: Reads selections from STYLESEED.md via --from-lock, with explicit CLI flags overriding lock values; --recipe auto and --palette auto map grammar to maintained defaults. - Drift detection: --check compares the current bundle hash against the manifest and exits non-zero when context has drifted, without rewriting files. - Use Case: Before building a SaaS dashboard UI, run the resolver with --grammar operations-console --adapter product-ui --domain saas --page dashboard so the agent reads one focused rule bundle instead of the full handbook. ## Quick Start Ask the agent to resolve the StyleSeed context from STYLESEED.md for your agent before starting UI work, then read the generated .styleseed/effective-rules.md.

Frequently Asked Questions about ss-resolve

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

FAQPage Schema
How do I generate a StyleSeed effective rules bundle?▼

Run scripts/resolve-context.mjs with --from-lock STYLESEED.md and --agent to compile the bundle. It writes .styleseed/effective-rules.md and a manifest.json recording selections and hashes. Explicit flags like --grammar and --adapter override lock values.

What is a StyleSeed output grammar and how do I choose one?▼

An output grammar is the functional visual language for the screen, such as operations-console, consumer-service, or editorial-reading. Run the resolver with --list to print all supported grammar, adapter, domain, page, recipe, palette, and profile IDs.

Can I use a project-local reference grammar instead of a built-in one?▼

Yes, pass reference:<slug> as the grammar and ensure .styleseed/rulesets/<slug>/RULESET.md exists in the project. Reference grammars also require a --fallback built-in grammar or a Grammar fallback value in STYLESEED.md.

How do I detect StyleSeed context drift without rewriting files?▼

Run the resolver with the --check flag. It recomputes the bundle hash and compares it to manifest.json, exiting with code 2 when the manifest is missing or the hash has drifted, and printing the current hash when everything matches.

What do --recipe auto and --palette auto do?▼

The auto options map the selected grammar to a maintained default brand recipe, then map that recipe to a contrast-verified semantic palette. Pass explicit recipe or palette IDs when the product needs a different morphology or color posture.

Related Skills

Explore More Design & Creative Skills→
styleframe-devstyleframe-dev
91

implement-recipe

Convert design.md files into TypeScript recipe implementations, barrel indices, and tests.

Official
Advanced
izzoaizzoa
22

ss-build

Build UI screens using StyleSeed rules with a lock-build-score-fix workflow.

Community
Advanced
sakhnenkoffsakhnenkoff
1

forge-judge

Evaluate Forge screens against the DESIGN.md contract and produce PASS/FAIL grades.

Community
Advanced
petrSimonidesXartpetrSimonidesXart

compose-team

Assemble per-sub-agent context bundles from evaluate-story output for deterministic spawning.

Community
Advanced