trellis-spec-bootstrap

Generates project-specific Trellis coding specs from real codebase analysis.

Updated Sep 5, 2026
One-click install
npx skills add https://github.com/jiozhaoyue/ST-BgLoader --skill trellis-spec-bootstrap-jiozhaoyue
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: trellis-spec-bootstrap
Source: https://github.com/jiozhaoyue/ST-BgLoader/tree/main/.cursor/skills/trellis-spec-bootstrap
Command: npx skills add https://github.com/jiozhaoyue/ST-BgLoader --skill trellis-spec-bootstrap-jiozhaoyue

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Teams adopting Trellis often end up with generic template specs that contain placeholder text and boilerplate advice unrelated to their actual codebase. This Skill bootstraps or refreshes .trellis/spec/ guidelines grounded in real source files, package boundaries, and local patterns. ## Core Features & Use Cases - Repository Analysis: Inspects architecture using GitNexus code knowledge graphs, ABCoder AST parsing, and direct source reads before writing any rule. - Spec Decomposition: Splits spec work by package, layer, or cross-cutting concern only when the codebase structure justifies it. - Evidence-Backed Writing: Produces spec files with concrete file paths, real examples, anti-patterns, and verification checks, removing all template placeholders. - Use Case: After initializing Trellis in a TypeScript monorepo, run this Skill to analyze each package and generate .trellis/spec/ docs that teach future agents the project's actual conventions. ## Quick Start Ask the agent to bootstrap the Trellis specs for this repository by analyzing the codebase and filling the .trellis/spec directory with project-specific guidance.

Frequently Asked Questions about trellis-spec-bootstrap

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

FAQPage Schema
How do I create Trellis specs for an existing codebase?▼

Initialize Trellis, then analyze the repository architecture using GitNexus, ABCoder, or direct source reads before writing specs. Decompose work by package or layer only where real boundaries exist, and fill each spec file with concrete patterns, file paths, and anti-patterns from the project.

What tools analyze a repository for spec generation?▼

GitNexus builds a code knowledge graph for module boundaries, execution flows, and impact analysis via `npx gitnexus analyze`. ABCoder parses code into UniAST for exact signatures and type shapes. Both expose MCP servers, but direct source reading also works.

Does this workflow require a specific AI agent or platform?▼

No, the workflow is platform-neutral and uses a single-agent model by default. GitNexus and ABCoder are optional tool choices configured through whatever MCP mechanism the host provides, and helper subagents are optional implementation details.

How do I verify Trellis specs have no placeholder text?▼

Run `grep -R "To be filled\|TODO: fill\|placeholder" .trellis/spec` as a final pass. Also check that index.md files match the actual spec file set, links resolve, and every rule is backed by a source file, test, or project document.

When should spec work be split into multiple tasks?▼

Split only along real ownership boundaries: one task per package with distinct conventions, per layer with different rules, or per cross-cutting pattern spanning packages. A small library usually needs one focused spec pass rather than artificial decomposition.