Skip to content

gustavo-meilus/superpipelines / plugins/superpipelines/skills/migrating-a-pipeline

migrating-a-pipeline

Verified

Migrate a legacy old-root Superpipelines workflow into the data-only pipeline layout.

fixtures/README.md

View on GitHub
# migrating-a-pipeline fixtures

Golden fixtures for the legacy→data-only reverse-translation contract.

## Legacy fixture note

Fixtures under `legacy-*` intentionally preserve old-root pipeline shapes so the migration workflow can prove lossless conversion into data-only artifacts. They are not examples for new pipeline scaffolding.

- `legacy-cc-pipeline/` — a minimal pre-v2 Claude Code pipeline (`doc-legacy`) as it would sit
  under a `.claude/`-era scope root: two agents (a worktree worker + a read-only Stage-1 reviewer),
  a legacy `run-doc-legacy` entry skill using direct `Task()`, a topology with no `agent_def`, and
  a legacy `superpipelines/registry.json` entry (`layout` implied legacy, no data root). This is the
  reverse-map (§3) + losslessness round-trip (§4) input.
- `migrated-expected/` — the expected data-only output: two CADs, the `entry.md` skeleton, and the
  data `registry.json` entry with `migrated_from` provenance. The losslessness golden.
- `legacy-injection-agent.md` — a legacy agent whose body carries a `<system-reminder>` injection
  tag; the Prompt-Injection Guardrail MUST refuse to migrate it and surface the path.

Expected per-agent verdicts (PHASE 1 analysis table):

| Legacy agent | CAD verdict | Why |
|---|---|---|
| `doc-writer` | `lossless-normalized` | `tools` incl. Write/Edit/Bash + `permissionMode: acceptEdits` + `isolation: worktree` → `write_files:true, run_shell:true, edit_tracked_source:true, isolation_required:true`; representation differs, effect identical (§3 rows). |
| `doc-reviewer` | `lossless-normalized` | `permissionMode: plan` + `disallowedTools: Write, Edit, Bash` + `tools: Read, Glob, Grep` → all caps `false`, `tool_hints.allow: [Read, Glob, Grep]`. |
| `legacy-injection-agent` | REFUSED | body carries `<system-reminder>` → guardrail halts and surfaces the path. |
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
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: 198kUpdated Claude CodeCodexVerifiedInstall
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: 119kUpdated Claude CodeCodexVerifiedInstall
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: 118kUpdated Claude CodeCodexVerifiedInstall