Complete SuperLoop v2 identity migration

.agents/plugins/marketplace.json

@@ -1,11 +1,11 @@
 {
-  "name": "loop-engineering",
+  "name": "superloop",
   "interface": {
-    "displayName": "Loop-Engineering"
+    "displayName": "SuperLoop"
   },
   "plugins": [
     {
-      "name": "loop-engineering",
+      "name": "superloop",
       "source": {
         "source": "url",
         "url": "./"

.claude-plugin/marketplace.json

@@ -1,20 +1,20 @@
 {
-  "name": "loop-engineering",
+  "name": "superloop",
   "owner": {
     "name": "sheepxux"
   },
   "metadata": {
-    "description": "Loop-Engineering Skill and local reference runtime",
-    "version": "1.1.0"
+    "description": "SuperLoop Skill and local reference runtime",
+    "version": "2.0.0"
   },
   "plugins": [
     {
-      "name": "loop-engineering",
-      "description": "Compile vague ideas into verified goals and governed agent loops.",
+      "name": "superloop",
+      "description": "SuperLoop compiles vague ideas into verified goals and governed agent loops.",
       "source": "./",
       "strict": true,
       "skills": [
-        "./skills/loop-engineering"
+        "./skills/superloop"
       ]
     }
   ]

.claude-plugin/plugin.json

@@ -1,16 +1,16 @@
 {
-  "name": "loop-engineering",
-  "description": "Complete Skill for compiling vague ideas into verified goals and governed durable agent loops.",
-  "version": "1.1.0",
+  "name": "superloop",
+  "description": "SuperLoop Skill for compiling vague ideas into verified goals and governed durable agent loops.",
+  "version": "2.0.0",
   "author": {
     "name": "sheepxux"
   },
-  "homepage": "https://github.com/sheepxux/Loop-Engineering",
-  "repository": "https://github.com/sheepxux/Loop-Engineering",
+  "homepage": "https://github.com/sheepxux/SuperLoop",
+  "repository": "https://github.com/sheepxux/SuperLoop",
   "license": "MIT",
   "keywords": [
     "agent-skills",
-    "loop-engineering",
+    "superloop",
     "strategy-evolution",
     "verification",
     "automation"

.codex-plugin/plugin.json

@@ -1,26 +1,26 @@
 {
-  "name": "loop-engineering",
-  "version": "1.1.0",
-  "description": "Turn vague ideas into verified goals and governed durable agent loops.",
+  "name": "superloop",
+  "version": "2.0.0",
+  "description": "SuperLoop turns vague ideas into verified goals and governed durable agent loops.",
   "author": {
     "name": "sheepxux",
     "url": "https://github.com/sheepxux"
   },
-  "homepage": "https://github.com/sheepxux/Loop-Engineering",
-  "repository": "https://github.com/sheepxux/Loop-Engineering",
+  "homepage": "https://github.com/sheepxux/SuperLoop",
+  "repository": "https://github.com/sheepxux/SuperLoop",
   "license": "MIT",
   "keywords": [
     "agent-skills",
-    "loop-engineering",
+    "superloop",
     "strategy-evolution",
     "verification",
     "automation"
   ],
   "skills": "./skills/",
   "interface": {
-    "displayName": "Loop-Engineering",
+    "displayName": "SuperLoop",
     "shortDescription": "Compile vague ideas into verified, durable agent loops",
-    "longDescription": "Use Loop-Engineering to turn a vague recurring or finite long-horizon idea into a verifiable Goal Contract and non-executable Loop Proposal, bind human approval to its exact digest, compile it deterministically, run bounded discovery or a fixed part-by-part Work Plan, and improve task strategy through matched benchmarks.",
+    "longDescription": "Use SuperLoop to turn a vague recurring or finite long-horizon idea into a verifiable Goal Contract and non-executable Loop Proposal, bind human approval to its exact digest, compile it deterministically, run bounded discovery or a fixed part-by-part Work Plan, and improve task strategy through matched benchmarks.",
     "developerName": "sheepxux",
     "category": "Developer Tools",
     "capabilities": [
@@ -35,7 +35,7 @@
       "Review this loop contract and find unsafe gaps.",
       "Build a benchmark-gated strategy evolution loop."
     ],
-    "websiteURL": "https://github.com/sheepxux/Loop-Engineering",
+    "websiteURL": "https://github.com/sheepxux/SuperLoop",
     "brandColor": "#000000",
     "logo": "./docs/assets/cover.png",
     "screenshots": []

.github/ISSUE_TEMPLATE/bug_report.md

@@ -25,4 +25,4 @@ labels: bug
 
 - Node:
 - OS:
-- `@sheepxux/loop-engineering` version:
+- `@sheepxux/superloop` version:

.gitignore

@@ -2,6 +2,6 @@ node_modules/
 dist/
 coverage/
 .DS_Store
-.loop-engineering/loops/*/runs/
+.superloop/loops/*/runs/
 *.log
 *.tgz

CHANGELOG.md

@@ -1,5 +1,16 @@
 # Changelog / 更新日志
 
+## v2.0.0 — SuperLoop identity and protocol v2 / SuperLoop 完整命名与 v2 协议
+
+- Renamed the npm package to `@sheepxux/superloop`, the canonical Skill and plugin ID to `superloop`, and the repository to `sheepxux/SuperLoop`.
+- Introduced `apiVersion: superloop/v2` across contracts, state, evaluator evidence, proposals, decisions, strategies, and experiments.
+- Moved the canonical runtime root to `.superloop/loops` and the administrative lock root to `.superloop-admin-locks`.
+- Renamed the canonical Skill directory to `skills/superloop` and updated Codex, Claude Code, ChatGPT, OpenClaw, and generic-harness adapters.
+- Treats the v1-to-v2 identity change as a contract migration: provenance-bound v1 contracts require a digest-linked proposal revision and fresh human approval; prior Part/Goal passes do not transfer automatically.
+- Updated every current schema ID, installation example, package fallback, generated template, example, test, and release manifest to the SuperLoop v2 identity.
+
+中文摘要：本版本完成 SuperLoop 的全量技术重命名：包名、Skill/插件标识、GitHub 地址、协议版本和持久状态目录全部进入新的正式命名。由于协议和路径属于获批契约的一部分，已有 v1 Loop 必须通过父提案相连的 revision、重新审批和新的 v2 初始化迁移，不能原地改写或自动继承既有通过状态。
+
 ## v1.1.0 — Idea-to-Loop design plane / 想法到循环设计层
 
 - Added a first-class `Propose` mode that turns vague recurring ideas into evidence-bearing Goal Contracts and non-executable `LoopProposal` artifacts.

CONTRIBUTING.md

@@ -1,6 +1,6 @@
 # Contributing
 
-Loop-Engineering is built around one rule: portable loop specs should be safer and clearer than ad hoc agent prompts.
+SuperLoop is built around one rule: portable loop specs should be safer and clearer than ad hoc agent prompts.
 
 ## Development