feat(M3): Shared-source generator for agent variants

This commit is contained in:
Stefano Fiorini
2026-05-03 21:09:22 -05:00
parent be993429c1
commit 86ad783f82
339 changed files with 20650 additions and 145 deletions
@@ -0,0 +1,19 @@
{
"$schema": "https://ai-coding-skills.dev/schemas/generated-manifest/v1.json",
"generator": "scripts/generate-skills.mjs",
"generatedRoot": "skills/do-task/opencode",
"files": [
{
"path": "SKILL.md",
"kind": "file",
"mode": "644",
"sha256": "ebbae7c050b3cfe0d1de8722cd71c8c10ba74de709adeec4ebb9645e2f770a03"
},
{
"path": "templates/task-plan.md",
"kind": "file",
"mode": "644",
"sha256": "a4c0ad9a9faaf738c503b403a285b6a005ee245b6b05f5af973c8e0224f6fefe"
}
]
}
+2
View File
@@ -3,6 +3,8 @@ name: do-task
description: Execute a single user-supplied prompt end-to-end with two reviewer loops (plan review + implementation review) in OpenCode. ALWAYS invoke when the user says `/do-task`, "do this task", "do task ...", "execute this task", or "make it so". Also invoke on the hint phrase "just do ...". Do NOT invoke on "implement this" (that phrase is reserved for implement-plan).
---
<!-- ⚠️ GENERATED FILE do not edit directly. Edit the canonical source in skills/do-task/_source/opencode/ and run `pnpm run sync:pi`. -->
# Do Task (OpenCode)
Execute an ad-hoc user prompt end-to-end: parse → clarify → plan (with reviewer loop) → implement (TDD-first where applicable) → verify → implementation review loop → commit → optional push → notify.
@@ -1,3 +1,4 @@
<!-- ⚠️ GENERATED FILE do not edit directly. Edit the canonical source in skills/do-task/_source/opencode/ and run `pnpm run sync:pi`. -->
# Task Plan: [Short Title]
> **Variant guardrail (OpenCode):** Sub-skills (`brainstorming`, `test-driven-development`, `verification-before-completion`, `finishing-a-development-branch`, `using-git-worktrees`) MUST be invoked through OpenCode's native skill tool from `~/.config/opencode/skills/superpowers/<skill>/SKILL.md`. Phase 1 MUST include the Bootstrap Superpowers Context step. Opencode reviewer calls MUST use `--agent plan` to stay read-only.