Skip to content

techygarg/lattice / skills/lattice-init

lattice-init

Verified

Guided setup and upgrade-check experience for Lattice projects -- scans the repository, detects existing configuration and outdated conventions, suggests refiners and available upgrades in priority order, and creates or reconciles the .lattice/ config. Bridges the gap between installing skills and getting first value, and between upgrading Lattice and adopting its newest conventions. Use when the user says 'lattice init', 'set up lattice', 'initialize lattice', 'get started with lattice', 'configure lattice for this project', 'check for lattice upgrades', or 'upgrade lattice conventions'.

FileKindSize
SKILL.mdskill md13.1 KB
references/requirements-migration.mdmarkdown3.8 KB

references/requirements-migration.md

View on GitHub
# Requirements Layout Migration

Read this file only when Step 3 of `lattice-init` has confirmed the user wants to migrate a legacy `.lattice/requirements/index.md` to the sharded layout. Follow these steps in order. This is a one-time, mechanical transform — no interview, no content judgment beyond the ambiguous-split case in Step 3 below.

## 1. Detect the plan

Read `.lattice/requirements/index.md`. For each `### [Epic Name]` section found, record: epic name, description (the paragraph before the feature table), and every feature row in its table.

For each feature row found, also open that feature's file and read its own `epic:` frontmatter. Record any disagreement between that value and the epic section the row was found under in the old index — do not resolve it here, just record it for Step 2.

## 2. Present the plan

Show the user:
- Epics detected, and the new file each will become (`.lattice/requirements/epics/{epic-slug}.md`)
- That `index.md` will be rewritten to the thin apex form
- That every feature file's `## Links` → `Epic index:` line will be repointed from `../index.md` to `../epics/{epic-slug}.md`
- That `.lattice/config.yaml` will gain `requirements_layout: sharded`
- **Any epic-placement disagreements found in Step 1** — for each, name the feature, its own `epic:` frontmatter value, and the epic section it was filed under in the old index. **STOP: ask the user to resolve each explicitly** — keep the feature file's `epic:` value (file it under that epic instead), or update the feature file's `epic:` to match its old placement. Do not guess or silently prefer one over the other.

**STOP: do not write anything until the user confirms this plan, including every epic-placement disagreement.**

## 3. Create epic files

For each detected epic, write `.lattice/requirements/epics/{epic-slug}.md` using the epic file template in `requirement-forge`'s `references/output-templates.md` — header (name, description, no `status:` field) plus the feature table seeded from the rows just extracted, using each feature's *resolved* epic assignment from Step 2. Keep only feature name and one-line summary in the table — drop any Status/Priority/Depends On values from the old rows; those fields live solely in each feature file's own frontmatter, untouched by this migration. Wrap the table in the generated-section boundary comments.

If the old `index.md` had a `## Source Materials` or `## Deferred Items` section whose content is scoped to one epic, move it into that epic's file. **STOP: if content spans multiple epics or the split is ambiguous, ask the user rather than guessing.**

## 4. Rewrite the apex index

Rewrite `.lattice/requirements/index.md` to the thin apex form (see `requirement-forge`'s `references/output-templates.md`): Definitions, Glossary if present, and a generated epic-list table pointing at the new epic files.

## 5. Fix back-links

For every file under `.lattice/requirements/features/`, update `## Links` → `Epic index:` to point at `../epics/{epic-slug}.md` instead of `../index.md`. For any feature where Step 2's disagreement resolution chose to update the feature's own `epic:` frontmatter, apply that change now — this is the one content edit beyond the link line, and only happens where the user explicitly confirmed it.

## 6. Write the config marker

Add `requirements_layout: sharded` to `.lattice/config.yaml`. Create the file if it does not exist; preserve all existing keys if it does.

## 7. Confirm

Summarize what changed: epics created, `index.md` rewritten, feature files repointed, config updated. No feature file's own content was modified beyond the one `Epic index:` link line — except where Step 2 surfaced an epic-placement disagreement and the user chose to update a feature's `epic:` frontmatter; list any such files explicitly here.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall