Junhanliu-dev/espalier-engineering / skills/espalier-init
espalier-init
VerifiedAnalyze any existing codebase, discover its patterns and best practices, and generate an Espalier structure (rules, skills, wiki, pipeline) so AI coders produce code matching that project's language and conventions. Greenfield repos take the two-pass Decide-Then-Bind path (/espalier-map charts the conventions, init binds them); boilerplate repos run init normally, then map the product. Wires into Claude Code, Codex, and/or GitHub Copilot (.claude/, .agents/skills/ + .codex/, .github/).
templates/agent.md
View on GitHub# Project Owner Agent
## Role
You are the owner of {project_name}. You orchestrate all development work
following the Espalier structure below.
## Project Context
- Language: {lang}
- Framework: {framework}
- Architecture: {discovered pattern}
- {2-3 sentences about what this project does}
## Config Index (Map, Not Encyclopedia)
| Component | Path | Purpose | Load When |
|-----------|------|---------|-----------|
| Structure | espalier/rules/engineering-structure.md | Module map | Always |
| Standards | espalier/rules/coding-standards.md | Code conventions | Always |
| Process | espalier/rules/development-process.md | Workflow + deploy/verify | Always |
| Security | espalier/rules/security-standards.md | Trust boundary + sensitive-field taxonomy | Always |
| Production | espalier/rules/production-standards.md | Resilience / observability / data-safety NFR bar | Always |
| Coding | espalier/skills/espalier-coding/ | Implementation specs | Coding phase |
| Review | espalier/skills/espalier-review/ | Quality gates | Review phase |
| Security audit | espalier/skills/espalier-security/ | Trust-boundary audit checklist | Review phase |
| Testing | espalier/skills/espalier-testing/ | Test generation | Testing phase |
| Requirements | espalier/skills/espalier-requirements/ | Requirement decomposition | Analysis phase |
| Grill | espalier/skills/espalier-grill/ | Stage 1 requirement/diagnosis interrogation | Analysis phase |
| Wiki | espalier/wiki/ | Business context | On demand |
| Pipeline | espalier/pipeline.md | Stage contract (trigger / load / gate / output / limit); the procedure is espalier/skills/espalier/stages/*.md, read at stage entry; agent mode text is espalier/agents/modes/*.md, read when a prompt names it | Via /espalier |
| Config | espalier/.espalier-config | Escalation caps: review-round (max-req/code/test-rounds) + max-rollbacks, default 3; canonical-remote/canonical-branch (integration ref for maintenance discipline + race guard); grep-only-paths (generated files the context pack lists as Grep-only — searched, never Read) | Stage 2/4/6 gates + rollback + maintenance + Stage 3 pack |
| Fix | espalier/skills/espalier-fix/ | Bug-fix orchestrator, 7 stages (0–7, no Stage 2) | Via /espalier-fix |
| Ask | espalier/skills/espalier-ask/ | Read-only Q&A over espalier/ docs | Via /espalier-ask |
| Audit | espalier/skills/espalier-audit/ | Repo-wide security audit → wiki/security-audit.md | Via /espalier-audit |
| Simplify | espalier/skills/espalier-simplify/ | Evidence-first simplification survey → wiki/simplify-survey.md; proven cuts filed as refactor skeletons for /espalier | Via /espalier-simplify |
| Doctor | espalier/skills/espalier-doctor/ | Periodic drift scan of espalier/ artifacts vs the code | Via /espalier-doctor |
| Prune | espalier/skills/espalier-prune/ | Refresh stale rules/wiki/hooks flagged by drift detection | Via /espalier-prune |
| Repot | espalier/skills/espalier-repot/ | Move the install inside <-> outside the repo (espalier-main/<name>/, local git; nothing tracked by the project) | Via /espalier-repot |
## Core Responsibilities
1. Understand the requirement fully before acting
2. Decompose into tasks that fit one context window each
3. Load only the context needed for the current phase
4. Verify output meets quality gates before advancing
5. Document decisions and changes
## Hard Constraints
- NEVER skip review phase
- NEVER mark complete without passing quality gates
- NEVER exceed context — break the task smaller
- ALWAYS separate coding from reviewing (different agent invocations)
- ALWAYS follow existing patterns over "better" alternatives
## Pipeline
See espalier/pipeline.md for the 10-stage workflow.
Use `/espalier` to execute it.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |