refactor: rename agent-zero to code-zero

.agents/skills/code-zero-architecture

@@ -0,0 +1 @@
+../../.skills/code-zero-architecture

.agents/skills/code-zero-cli

@@ -0,0 +1 @@
+../../.skills/code-zero-cli

.agents/skills/code-zero-safety

@@ -0,0 +1 @@
+../../.skills/code-zero-safety

.claude/skills/code-zero-architecture

@@ -0,0 +1 @@
+../../.agents/skills/code-zero-architecture

.claude/skills/code-zero-cli

@@ -0,0 +1 @@
+../../.agents/skills/code-zero-cli

.claude/skills/code-zero-safety

@@ -0,0 +1 @@
+../../.agents/skills/code-zero-safety

.agent-zero.example.yml → .code-zero.example.yml

@@ -14,8 +14,8 @@ proactive:
 # committed to.
 issues:
   enabled: false
-  requireLabel: agent-zero
-  branchPrefix: agent-zero/
+  requireLabel: code-zero
+  branchPrefix: code-zero/
   # Report the validation verdict back on the issue as a comment: confirmed with evidence, not
   # confirmed with the rejection reasons, or inconclusive for a human. Report-only.
   validationComment: true
@@ -27,7 +27,7 @@ checks: []
 
 autofix:
   enabled: false
-  # Confidence required before Agent Zero may change files.
+  # Confidence required before Code Zero may change files.
   minConfidence: 0.85
   # mechanical is the conservative default. behavioral may be added explicitly; high-impact
   # changes always require human approval and cannot be enabled here.
@@ -65,7 +65,7 @@ runner:
   # cpus: '2'
   # memory: 4g
   # Pre-provisioned network used when permissions.network is restricted.
-  # network: agent-zero
+  # network: code-zero
   maxOutputBytes: 200000
 
 model:
@@ -75,7 +75,7 @@ model:
   #
   # claude-code and codex-cli spend a Claude Pro/Max or ChatGPT Plus/Pro subscription through the
   # vendor CLI installed on the host. They take no credential and stay off until the operator sets
-  # AGENT_ZERO_ENABLE_CLAUDE_CODE_PROVIDER or AGENT_ZERO_ENABLE_CODEX_CLI_PROVIDER to true, because
+  # CODE_ZERO_ENABLE_CLAUDE_CODE_PROVIDER or CODE_ZERO_ENABLE_CODEX_CLI_PROVIDER to true, because
   # the session is bound to one host and one personal account.
   provider: openai-compatible
   name: gpt-5

.env.example

@@ -11,7 +11,7 @@
 #   .env                     this file — the `zero` CLI: model providers,
 #                            subscription transports, GITHUB_TOKEN
 #   apps/dashboard/.env      the dashboard process, loaded by Nuxt itself:
-#                            control plane (AGENT_ZERO_CONTROL_PLANE_*),
+#                            control plane (CODE_ZERO_CONTROL_PLANE_*),
 #                            webhook ingress, DATABASE_URL, authentication,
 #                            mail, NUXT_PUBLIC_SITE_URL
 #   apps/marketing/.env      the marketing site, loaded by Nuxt itself:
@@ -25,39 +25,39 @@
 # See docs: apps/docs/guide/environment-variables.md.
 
 OPENAI_API_KEY=
-AGENT_ZERO_MODEL=gpt-5
+CODE_ZERO_MODEL=gpt-5
 GITHUB_TOKEN=
 
 # Subscription model transports (model.provider: claude-code / codex-cli). These take no API key:
 # they drive the vendor CLI already logged in on this host, so they are single-tenant and bound to
 # this machine. Each stays off unless its flag is exactly "true", so a CI runner without the CLI
 # installed never spawns a subprocess.
-AGENT_ZERO_ENABLE_CLAUDE_CODE_PROVIDER=false
-AGENT_ZERO_ENABLE_CODEX_CLI_PROVIDER=false
+CODE_ZERO_ENABLE_CLAUDE_CODE_PROVIDER=false
+CODE_ZERO_ENABLE_CODEX_CLI_PROVIDER=false
 # Absolute path to the executable when it is not on PATH.
-AGENT_ZERO_CLAUDE_CODE_PATH=
-AGENT_ZERO_CODEX_PATH=
-# Required when runner.isolation is "container" in .agent-zero.yml. Without it, claude-code is
+CODE_ZERO_CLAUDE_CODE_PATH=
+CODE_ZERO_CODEX_PATH=
+# Required when runner.isolation is "container" in .code-zero.yml. Without it, claude-code is
 # refused rather than spawned unisolated on the control-plane host: this image must have the
 # `claude` CLI installed, and only claude-code is affected (codex-cli's vendor SDK has no spawn
 # hook to route through a container either way; see README "Known limits"). The container gets no
 # repository checkout and unrestricted network — both deliberately different from runner.image and
 # permissions.network, which govern an untrusted checkout's own commands, not this process.
-AGENT_ZERO_CLAUDE_CODE_CONTAINER_IMAGE=
+CODE_ZERO_CLAUDE_CODE_CONTAINER_IMAGE=
 # Executable name/path to run inside that container. Defaults to "claude"; only needs setting if
 # the image installs it somewhere not on PATH. The vendor SDK's own resolved path (its bundled
-# native binary, or AGENT_ZERO_CLAUDE_CODE_PATH) is a host path and never used inside the container.
-AGENT_ZERO_CLAUDE_CODE_CONTAINER_EXECUTABLE=
+# native binary, or CODE_ZERO_CLAUDE_CODE_PATH) is a host path and never used inside the container.
+CODE_ZERO_CLAUDE_CODE_CONTAINER_EXECUTABLE=
 # Host directory holding the CLI's own `claude login` session, mounted read-only into that
 # container along with its sibling ~/.claude.json. Defaults to ~/.claude; only needs setting if the
 # CLI was configured with a custom CLAUDE_CONFIG_DIR on this host.
 CLAUDE_CONFIG_DIR=
 # Longest a run may block in total waiting for a spent subscription window to reopen, before
 # resuming the interrupted CLI session. Default 3600000 (one hour); 0 disables waiting so a run
 # fails straight onto the fallback below. A reset further out than this is never waited on.
-AGENT_ZERO_SUBSCRIPTION_LIMIT_WAIT_MS=3600000
+CODE_ZERO_SUBSCRIPTION_LIMIT_WAIT_MS=3600000
 # Optional API-key transport a subscription run degrades to when the CLI is missing, its session
 # expired, or its usage window is spent and too far from reopening to wait out. Both must be set
 # together, and the provider must not itself be subscription-based.
-AGENT_ZERO_MODEL_FALLBACK_PROVIDER=
-AGENT_ZERO_MODEL_FALLBACK_MODEL=
+CODE_ZERO_MODEL_FALLBACK_PROVIDER=
+CODE_ZERO_MODEL_FALLBACK_MODEL=

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -5,7 +5,7 @@ labels: ['bug']
 body:
   - type: markdown
     attributes:
-      value: Thanks for helping improve Agent Zero. Do not include secrets or private repository content.
+      value: Thanks for helping improve Code Zero. Do not include secrets or private repository content.
   - type: checkboxes
     id: existing
     attributes:

.github/ISSUE_TEMPLATE/config.yml

@@ -1,7 +1,7 @@
 blank_issues_enabled: false
 contact_links:
   - name: Security report
-    url: https://github.com/wolfstar-project/agent-zero/security/advisories/new
+    url: https://github.com/wolfstar-project/code-zero/security/advisories/new
     about: Report vulnerabilities privately. Do not open a public issue.
   - name: Questions and discussion
     url: https://join.wolfstar.rocks

.github/copilot-instructions.md

@@ -1,4 +1,4 @@
-# Agent Zero repository instructions
+# Code Zero repository instructions
 
 Read `/AGENTS.md` before editing and load the relevant skill from `/.agents/skills/`.
 

.github/workflows/release.yaml

@@ -35,5 +35,5 @@ jobs:
         run: |
           node --input-type=module --eval "Promise.all([import('./packages/shared/dist/index.mjs'), import('./packages/shared/package.json', { with: { type: 'json' } })]).then(([built, pkg]) => { if (built.version !== pkg.default.version) { console.error('Injected version', built.version, 'does not match package.json version', pkg.default.version); process.exit(1) } })"
 
-      # Publishing remains intentionally absent until the @agent-zero package
+      # Publishing remains intentionally absent until the @code-zero package
       # scope and npm trusted publishing policy are configured.

.github/workflows/semantic-pull-requests.yml

@@ -19,7 +19,7 @@ jobs:
       contents: read
       pull-requests: read # analyze pull requests
       statuses: write # publish validation status
-    if: github.repository == 'wolfstar-project/agent-zero'
+    if: github.repository == 'wolfstar-project/code-zero'
     runs-on: ubuntu-latest
     name: 🏷️ Validate PR title
     steps:

.gitignore

@@ -16,7 +16,7 @@ test-report.junit.xml
 .env
 .env.*
 !.env.example
-.agent-zero/
+.code-zero/
 .data/
 *.log
 .DS_Store

.skills/agent-zero-architecture/SKILL.md → .skills/code-zero-architecture/SKILL.md

@@ -1,9 +1,9 @@
 ---
-name: agent-zero-architecture
-description: Use when adding features, moving code, or changing dependencies across Agent Zero packages and adapters.
+name: code-zero-architecture
+description: Use when adding features, moving code, or changing dependencies across Code Zero packages and adapters.
 ---
 
-# Agent Zero architecture
+# Code Zero architecture
 
 Keep dependency direction explicit while changing the monorepo.
 
@@ -36,12 +36,12 @@ Keep dependency direction explicit while changing the monorepo.
 - `agent`: orchestration, the lifecycle machine, and the validation policy.
 - `cli`: argument parsing and terminal presentation. Composition-root glue that decides *how* to
   isolate the `claude-code` CLI process lives here (`subscription-isolation.ts`, duplicated
-  identically in `api` rather than shared — it needs both `AgentZeroConfig` and an operator
+  identically in `api` rather than shared — it needs both `CodeZeroConfig` and an operator
   environment variable, which neither `models` nor `runner` should own). Refuses the transport
   outright when `runner.isolation: container` is declared but no CLI container image is configured,
   rather than falling back to an unisolated host spawn — reported to `modelFromEnvironment` as a
   refusal reason, not by disabling the enable flag, so a configured
-  `AGENT_ZERO_MODEL_FALLBACK_PROVIDER` still gets a turn. `api` applies the same refusal-reason
+  `CODE_ZERO_MODEL_FALLBACK_PROVIDER` still gets a turn. `api` applies the same refusal-reason
   mechanism for a `RunnerPool` lease, which cannot route the CLI's duplex spawn through a
   `SandboxProvider`'s bounded `Runner` contract.
 - `database`: Postgres schema, the Drizzle client factory, and checked-in migrations. No policy, and the only package that names a table or opens a connection.

.skills/agent-zero-cli/SKILL.md → .skills/code-zero-cli/SKILL.md

@@ -1,9 +1,9 @@
 ---
-name: agent-zero-cli
-description: Use when adding or changing Agent Zero CLI commands, flags, prompts, output, exit codes, or version display.
+name: code-zero-cli
+description: Use when adding or changing Code Zero CLI commands, flags, prompts, output, exit codes, or version display.
 ---
 
-# Agent Zero CLI
+# Code Zero CLI
 
 The CLI uses `@bomb.sh/args` for typed parsing and `@clack/prompts` for human-friendly interaction.
 
@@ -18,7 +18,7 @@ The CLI uses `@bomb.sh/args` for typed parsing and `@clack/prompts` for human-fr
 - Obtain the displayed version through the injected shared version marker.
 - CLI commands request runtime behavior through typed APIs; they do not execute repository commands directly.
 - A session obtained by `zero login` is stored per deployment origin in
-  `$XDG_CONFIG_HOME/agent-zero/credentials.json`, owner-readable only (`packages/cli/src/credentials.ts`).
+  `$XDG_CONFIG_HOME/code-zero/credentials.json`, owner-readable only (`packages/cli/src/credentials.ts`).
   Never write a token beside the checkout, never log one, and never key the store by anything but
   the normalized origin — one workstation is expected to hold a cloud-managed and a self-hosted
   session at once.
@@ -33,4 +33,4 @@ The CLI uses `@bomb.sh/args` for typed parsing and `@clack/prompts` for human-fr
 3. Add Clack presentation only after parsing succeeds.
 4. Document the command in `README.md` and `CONTRIBUTING.md` when contributor-facing.
 5. Verify interactive and non-interactive behavior.
-6. Run `aube run test --filter @agent-zero/cli` and the root checks.
+6. Run `aube run test --filter @code-zero/cli` and the root checks.

.skills/agent-zero-safety/SKILL.md → .skills/code-zero-safety/SKILL.md

@@ -1,9 +1,9 @@
 ---
-name: agent-zero-safety
+name: code-zero-safety
 description: Use for runner changes, autonomous state transitions, repository writes, command execution, secrets, or observe/fix policy.
 ---
 
-# Agent Zero safety
+# Code Zero safety
 
 Safety properties are behavior, not documentation. Back every change with deterministic checks.
 

.skills/orpc-server/SKILL.md

@@ -68,7 +68,7 @@ description: Use when changing apps/dashboard server routes, oRPC contracts, han
   taking the environment record as an argument rather than reading `process.env` themselves. They
   deliberately do not move to Nuxt's `runtimeConfig`: its defaults are baked at build time and
   would require renaming every variable to a `NUXT_`-prefixed form, while a deployment sets
-  `GITHUB_WEBHOOK_SECRET` and `AGENT_ZERO_CHECKOUT_PATH` at run time.
+  `GITHUB_WEBHOOK_SECRET` and `CODE_ZERO_CHECKOUT_PATH` at run time.
 - Keep transport-specific headers, status mapping, and request objects out of runtime packages.
 - Nuxt's Nitro server is the only top-level HTTP host; do not introduce Express, Hono, or a second
   one.
@@ -86,5 +86,5 @@ description: Use when changing apps/dashboard server routes, oRPC contracts, han
 3. Keep the handler thin: validate, authorize, delegate, translate.
 4. Add router tests with `createRouterClient`, without opening a real network port.
 5. Update the README client example when the public router shape changes.
-6. Run `aube run test --filter @agent-zero/api`, then `--filter @agent-zero/dashboard`, typecheck,
+6. Run `aube run test --filter @code-zero/api`, then `--filter @code-zero/dashboard`, typecheck,
    and build. For KV or transport-mounting changes, also run a real `nuxt build` and hit the route.

.skills/tsdown/SKILL.md

@@ -1,9 +1,9 @@
 ---
 name: tsdown
-description: Build or change Agent Zero packages and apps with the repository's shared tsdown configuration, package validation, and version injection.
+description: Build or change Code Zero packages and apps with the repository's shared tsdown configuration, package validation, and version injection.
 ---
 
-# tsdown in Agent Zero
+# tsdown in Code Zero
 
 Use this skill when changing build entries, output formats, declarations, package exports, executable output, or version injection.