Phase 3 status: complete.
This matrix maps the pinned upstream repository at commit 984023ddac0d5e27624f2baacde6105e45de375f to the corrected Codex-native architecture. It supersedes the initial Phase 2 matrix.
Verified upstream file coverage ledger: 417 files total.
| Source category | Count | Disposition |
|---|---|---|
.claude/agents/*.md |
49 | Port to Codex custom agents |
.claude/skills/*/SKILL.md |
73 | Port to Codex repo skills with upstream names |
.claude/hooks/* |
12 | Port 11 lifecycle hooks; replace notify.sh behavior with Codex native notification guidance |
.claude/rules/*.md |
11 | Port path-scoped guidance into the root-routed .codex/instructions/path-rules/ library; keep command policy separate |
.claude/docs/templates/** |
40 | Copy to .codex/docs/templates/** |
.claude/docs/** excluding templates |
23 | Copy or rewrite to neutral Codex docs |
.claude/settings.json, .claude/statusline.sh |
2 | Replace with Codex config/rules/hooks/native status-line/status skill |
.claude/agent-memory/lead-programmer/MEMORY.md |
1 | Port to repo-local Codex agent memory and bind from custom agent instructions |
| Root files | 7 | Port docs/license where relevant; replace CLAUDE.md with AGENTS.md block |
.github/** |
5 | Copy as optional distribution/community docs only |
CCGS Skill Testing Framework/** |
127 | Copy to Codex testing framework docs |
docs/** |
62 | Preserve shared neutral docs; route documentation guidance through root AGENTS.md to docs-directory.md |
design/** |
2 | Preserve shared registry; route design guidance through root AGENTS.md to design-directory.md |
src/** |
2 | Preserve placeholder; route source guidance through root AGENTS.md to source-code.md and matching specialist rules |
production/** |
1 | Preserve shared neutral state placeholder |
Disposition key:
ported: converted into a Codex runtime or documentation asset.replaced: behavior retained through a different Codex-native mechanism.shared: retained at the same neutral project path.not applicable: intentionally omitted with rationale.blocked: cannot be completed without new data or a user decision.
Parity key:
exact: same behavior and same or equivalent native capability.semantic equivalent: same workflow/artifact behavior through a different Codex primitive.partial: important behavior preserved, but some Claude feature has no exact Codex equivalent.not applicable: intentionally omitted with rationale.blocked: cannot be mapped without a new capability or user decision.
| Codex primitive | Verified target path or mechanism | Notes |
|---|---|---|
| Project instructions | Root AGENTS.md router plus .codex/instructions/path-rules/*.md; framework-local AGENTS.md only where separately useful |
Codex loads the root-to-CWD instruction chain at session start, not rules selected per edited file. The shipped router is therefore the canonical CCGS path-rule mechanism; nested files are optional supplements, not parity enforcement. |
| Repo skills | .agents/skills/<skill>/SKILL.md |
Official docs describe direct skill folders under .agents/skills. Use upstream folder/name values for CCGS parity and add collision auditing instead of relying on undocumented suppression. |
| Custom subagents | .codex/agents/<agent>.toml |
Official docs describe standalone TOML files under .codex/agents. Preserve upstream hyphenated role names for filenames and name fields; add collision auditing rather than artificial prefixes. |
| Hooks | .codex/hooks.json plus .codex/hooks/* scripts |
Project hooks require project trust. Supported lifecycle events do not include Claude's Notification hook event. |
| Command rules | .codex/rules/settings.rules |
Codex command approval rules, not path-scoped content rules. |
| Permission profiles | .codex/config.toml with default_permissions and [permissions.<name>] |
Select one complete native profile without a project-local approval_policy or legacy sandbox keys. |
| TUI status line | .codex/config.toml [tui].status_line |
Verified Codex-native footer configuration for built-in items such as model/context/current-dir/git. No verified project script item equivalent to Claude statusLine.command. |
| Notifications | user-level notify, [tui].notifications, [tui].notification_method, [tui].notification_condition |
Codex has native notification settings, but project-local config cannot set notify; document user-level setup instead of installing notify.sh as a hook. |
| Plugins | Optional .codex-plugin/plugin.json package |
Optional later distribution mechanism. Not the base architecture because loose repo files are simpler for coexistence. |
| MCP elicitation | Optional MCP tool prompt/elicitation support | Only use if a real MCP server is introduced. Default plan rewrites AskUserQuestion as numbered natural-language choices. |
Official docs used: https://developers.openai.com/codex/guides/agents-md, https://developers.openai.com/codex/skills, https://developers.openai.com/codex/subagents, https://developers.openai.com/codex/hooks, https://developers.openai.com/codex/permissions, https://developers.openai.com/codex/rules, https://developers.openai.com/codex/plugins, https://developers.openai.com/codex/config-reference.
| Upstream path/component | Upstream behavior | Claude Code mechanism | Codex target mechanism | Target path/name | Disposition | Parity | Implementation notes | Verification method |
|---|---|---|---|---|---|---|---|---|
CLAUDE.md |
Root studio operating instructions, lifecycle guidance, startup nudge | Root Claude instruction file with imported docs | Marker-managed root Codex instructions | AGENTS.md CCGS block |
replaced | semantic equivalent | Inline critical guidance; remove .claude/, CLAUDE.md, and bare Claude slash-command references; tell users to run start. |
Grep runtime docs for forbidden refs; wc -c AGENTS.md; prompt-input/manual instruction check. |
src/CLAUDE.md |
Source-code scope guidance | Nested Claude instructions | Root-routed Codex path guidance | .codex/instructions/path-rules/source-code.md plus specialist rule files |
replaced | partial | Root AGENTS.md requires reading the matching rules before edits; this is advisory rather than upstream path-trigger enforcement. |
Router-chain validation and instruction-budget check. |
design/CLAUDE.md |
Design-document scope guidance | Nested Claude instructions | Root-routed Codex path guidance | .codex/instructions/path-rules/design-directory.md |
replaced | partial | Preserve design and registry guidance; keep design/registry/entities.yaml shared. |
Router-chain validation; no runtime .claude refs. |
docs/CLAUDE.md |
Documentation scope guidance | Nested Claude instructions | Root-routed Codex path guidance | .codex/instructions/path-rules/docs-directory.md |
replaced | partial | Preserve docs standards without depending on Claude imports. | Router-chain validation; instruction-budget check. |
CCGS Skill Testing Framework/CLAUDE.md |
Test-framework specific guidance | Nested Claude instructions | Testing-framework docs/instructions | CCGS Skill Testing Framework/AGENTS.md or README section |
replaced | semantic equivalent | Preserve the upstream framework path; replace Claude-specific guidance with Codex guidance. | File existence; no runtime .claude dependency. |
.claude/settings.json permissions |
Allow/deny command and path policy | Claude permissions in settings JSON | Complete Codex permission profile plus command rules | .codex/config.toml, .codex/rules/settings.rules |
replaced | semantic equivalent | Select [permissions.game_studios] extends = ":workspace" through default_permissions, restore writes to .git, .agents, and .codex, deny sensitive env paths, and allow only github.com; command policies remain in .rules. Do not set project-local approval_policy: the completed profile must support normal handoff without escalation. |
TOML parse; rules lint; forbidden command fixture; fresh trusted session proves the profile is active with Git/runtime writes, .env* denials, and GitHub network access; end-to-end handoff proves commit and push. |
.claude/settings.json hooks |
Hook event wiring | Claude settings hook blocks | Codex hooks config | .codex/hooks.json |
replaced | semantic equivalent | Map supported lifecycle events only. Do not include Claude's Notification event; map notification behavior separately through user-level Codex notification docs. |
JSON parse; event allowlist; command-path existence. |
.claude/statusline.sh |
Always-on context/model/stage/session breadcrumb | Claude scriptable status line command | Codex native footer config plus startup/on-demand studio status | .codex/config.toml [tui].status_line, .codex/hooks/studio-status-on-start.sh, .agents/skills/studio-status/SKILL.md |
replaced | partial; Stage: footer blocked |
Preserve context/model visibility with Codex built-in status_line items. Preserve stage/review/session breadcrumb through shared-state parsing in studio-status-on-start.sh and studio-status; actual footer Stage: parity is blocked until Codex exposes a documented project custom footer item. |
Config parse; status_line item allowlist; hook fixture; studio-status smoke reads production/session-state/active.md; no fake custom footer key. |
.claude/agent-memory/lead-programmer/MEMORY.md |
Lead-programmer project memory seed | Claude agent memory file plus memory: project metadata |
Repo-local Codex memory contract read by custom agent instructions | .codex/agent-memory/lead-programmer/MEMORY.md plus .codex/agents/lead-programmer.toml binding |
ported | semantic equivalent | Rewrite Claude skill paths to .agents/skills/<name>/SKILL.md and keep canonical paths. Agent TOML must instruct lead-programmer to read the memory file before skill authoring, path canonicalization, and convention work. Do not write ~/.codex/memories. |
File existence; TOML binding scan; static scan: no writes to ~/.codex/memories; content rewrite scan. |
production/session-state/.gitkeep |
Keeps shared session-state directory in template | Neutral placeholder | Shared neutral state | same path | shared | exact | Installer may create directory if absent, but uninstaller must not remove shared state. | Install/uninstall fixture; shared-state hash unchanged except intentional writes. |
design/registry/entities.yaml |
Entity registry used by design consistency workflows | Neutral YAML data | Shared neutral data | same path | shared | exact | Codex and Claude can both read/write this. | YAML parse; coexistence write/read smoke. |
src/.gitkeep |
Placeholder source directory | Neutral placeholder | Shared neutral placeholder | same path | shared | exact | Do not remove on uninstall. | Install/uninstall fixture. |
The installed Codex hooks are upstream behavior with Codex adapter edits, not
freeform rewrites. Adapter edits are limited to Codex-owned paths, $skill
invocation text, CCGS_ROOT root-stable execution, git -C "$ccgs_root" calls,
and Codex hook payload parsing through .codex/lib/hooks.sh.
| Upstream hook | Upstream behavior | Claude event | Codex target | Disposition | Parity | Implementation notes | Verification method |
|---|---|---|---|---|---|---|---|
.claude/hooks/session-start.sh |
Session startup context | SessionStart |
.codex/hooks/session-start.sh |
ported | semantic equivalent | Rewrite stdin JSON fields to Codex hook schema. | Fixture stdin test; hooks.json event check. |
.claude/hooks/detect-gaps.sh |
Startup gap detection | SessionStart |
.codex/hooks/detect-gaps.sh |
ported | semantic equivalent | Must read neutral production/, design/, docs/architecture/ paths. |
Fixture with missing/complete project state. |
.claude/hooks/validate-commit.sh |
Git commit gate | PreToolUse on Bash |
.codex/hooks/validate-commit.sh plus .codex/rules/settings.rules |
ported | semantic equivalent | Blocking command policy belongs primarily in .rules; hook can add context-specific checks. |
Bash fixture; command-rule match/not_match tests. |
.claude/hooks/validate-push.sh |
Git push protected-branch warning | PreToolUse on Bash |
.codex/hooks/validate-push.sh plus .codex/rules/settings.rules |
ported | semantic equivalent | Warn, do not block, for explicit or current-branch pushes to develop, main, or master. |
Bash fixture; protected push advisory smoke. |
.claude/hooks/validate-assets.sh |
Asset validation after edits | PostToolUse on Write/Edit |
.codex/hooks/validate-assets.sh |
ported | semantic equivalent | Parse Codex apply_patch payload paths from tool_input.command with legacy tool_input.patch fallback; keep naming advisory and invalid JSON feedback after the edit. Exit 2 on PostToolUse reports feedback but does not roll back side effects. |
apply_patch command-field fixture with naming warning and invalid JSON feedback; legacy patch fallback assertion. |
.claude/hooks/validate-skill-change.sh |
Skill metadata/content guard | PostToolUse on Write/Edit |
.codex/hooks/validate-skill-change.sh |
ported | semantic equivalent | Parse Codex apply_patch payload paths from tool_input.command with legacy fallback and advise $skill-test static <name> for .agents/skills/<name>/SKILL.md after the edit. |
Fixture modifies sample skill through command-field payload; lint output. |
.claude/hooks/pre-compact.sh |
Preserve session notes before compaction | PreCompact |
.codex/hooks/pre-compact.sh |
ported | semantic equivalent | Codex matcher values are manual/auto. | PreCompact fixture. |
.claude/hooks/post-compact.sh |
Reload session notes after compaction | PostCompact |
.codex/hooks/post-compact.sh |
ported | semantic equivalent | Read/write only Codex-owned logs plus shared neutral handoff docs. | PostCompact fixture. |
.claude/hooks/log-agent.sh |
Log agent start | SubagentStart |
.codex/hooks/log-agent.sh |
ported | semantic equivalent | Codex fields include agent_id and agent_type. |
SubagentStart fixture; log assertion. |
.claude/hooks/log-agent-stop.sh |
Log agent stop | SubagentStop |
.codex/hooks/log-agent-stop.sh |
ported | semantic equivalent | Same log namespace as start. | SubagentStop fixture; log assertion. |
.claude/hooks/session-stop.sh |
End-session summary/logging | Stop |
.codex/hooks/session-stop.sh |
ported | semantic equivalent | Use common fields only; no Claude matcher assumptions. | Stop fixture. |
.claude/hooks/notify.sh |
Desktop notification when Claude emits a notification | Notification hook event |
Native Codex notifications documented for user-level setup; no project hook | .codex/docs/COEXISTENCE.md / .codex/docs/README.md notification section |
replaced | partial | Codex has native TUI notifications and user-level notify, but no Claude-style Notification hook event, and project config cannot set user-level notify. Upstream script is Windows-oriented, so the faithful Codex port documents native notification setup rather than installing this script as a project hook. |
| n/a | Codex startup status summary | SessionStart |
.codex/hooks/studio-status-on-start.sh |
added | Codex-additive | Not upstream parity; preserves status visibility using Codex-native shared state reads. | Fixture stdin test; documentation marks additive. |
Upstream .claude/rules/*.md files are path-scoped instruction rules. CCGS
ships them as .codex/instructions/path-rules/*.md and makes root AGENTS.md
the canonical router. Codex's nested instruction chain is selected from the
session working directory, not dynamically for every edited file, so nested
AGENTS.md files do not restore upstream path-trigger semantics for a
root-launched session. This is an intentional, advisory partial-parity mapping.
Codex command .rules remain reserved for shell command approval policy.
| Upstream rule | Codex target | Disposition | Parity | Verification |
|---|---|---|---|---|
.claude/rules/ai-code.md |
.codex/instructions/path-rules/ai-code.md |
replaced | partial | Root-router chain validation. |
.claude/rules/data-files.md |
.codex/instructions/path-rules/data-files.md |
replaced | partial | Root-router chain validation. |
.claude/rules/design-docs.md |
.codex/instructions/path-rules/design-docs.md |
replaced | partial | Root-router chain validation. |
.claude/rules/engine-code.md |
.codex/instructions/path-rules/engine-code.md |
replaced | partial | Root-router chain validation. |
.claude/rules/gameplay-code.md |
.codex/instructions/path-rules/gameplay-code.md |
replaced | partial | Root-router chain validation. |
.claude/rules/narrative.md |
.codex/instructions/path-rules/narrative.md |
replaced | partial | Root-router chain validation. |
.claude/rules/network-code.md |
.codex/instructions/path-rules/network-code.md |
replaced | partial | Root-router chain validation. |
.claude/rules/prototype-code.md |
.codex/instructions/path-rules/prototype-code.md |
replaced | partial | Root-router chain validation. |
.claude/rules/shader-code.md |
.codex/instructions/path-rules/shader-code.md |
replaced | partial | Root-router chain validation. |
.claude/rules/test-standards.md |
.codex/instructions/path-rules/test-standards.md |
replaced | partial | Root-router chain validation. |
.claude/rules/ui-code.md |
.codex/instructions/path-rules/ui-code.md |
replaced | partial | Root-router chain validation. |
Common mapping for all rows: upstream behavior is a role persona for delegation; Claude mechanism is a .claude/agents/*.md Task subagent; Codex mechanism is a project custom agent TOML. Tool fences, memory, maxTurns, and worktree isolation are partial and covered in the risk register.
| Upstream agent | Model tier | Codex target file | Codex agent name | Disposition | Parity | Notes | Verification |
|---|---|---|---|---|---|---|---|
.claude/agents/accessibility-specialist.md |
sonnet | .codex/agents/accessibility-specialist.toml |
accessibility-specialist |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/ai-programmer.md |
sonnet | .codex/agents/ai-programmer.toml |
ai-programmer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/analytics-engineer.md |
sonnet | .codex/agents/analytics-engineer.toml |
analytics-engineer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/art-director.md |
sonnet | .codex/agents/art-director.toml |
art-director |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/audio-director.md |
sonnet | .codex/agents/audio-director.toml |
audio-director |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/community-manager.md |
haiku | .codex/agents/community-manager.toml |
community-manager |
ported | semantic equivalent | model map: gpt-5.4-mini, low |
TOML parse; field check; role smoke. |
.claude/agents/creative-director.md |
opus | .codex/agents/creative-director.toml |
creative-director |
ported | partial | preserve pinned skills brainstorm, design-review as guidance; model map gpt-5.5, high |
TOML parse; pinned-skill guidance scan; role smoke. |
.claude/agents/devops-engineer.md |
haiku | .codex/agents/devops-engineer.toml |
devops-engineer |
ported | semantic equivalent | model map: gpt-5.4-mini, low |
TOML parse; field check; role smoke. |
.claude/agents/economy-designer.md |
sonnet | .codex/agents/economy-designer.toml |
economy-designer |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/engine-programmer.md |
sonnet | .codex/agents/engine-programmer.toml |
engine-programmer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/game-designer.md |
sonnet | .codex/agents/game-designer.toml |
game-designer |
ported | partial | preserve pinned skills design-review, balance-check, brainstorm as guidance |
TOML parse; pinned-skill guidance scan; role smoke. |
.claude/agents/gameplay-programmer.md |
sonnet | .codex/agents/gameplay-programmer.toml |
gameplay-programmer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/godot-csharp-specialist.md |
sonnet | .codex/agents/godot-csharp-specialist.toml |
godot-csharp-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/godot-gdextension-specialist.md |
sonnet | .codex/agents/godot-gdextension-specialist.toml |
godot-gdextension-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/godot-gdscript-specialist.md |
sonnet | .codex/agents/godot-gdscript-specialist.toml |
godot-gdscript-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/godot-shader-specialist.md |
sonnet | .codex/agents/godot-shader-specialist.toml |
godot-shader-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/godot-specialist.md |
sonnet | .codex/agents/godot-specialist.toml |
godot-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/lead-programmer.md |
sonnet | .codex/agents/lead-programmer.toml |
lead-programmer |
ported | partial | preserve pinned skills code-review, architecture-decision, tech-debt; bind .codex/agent-memory/lead-programmer/MEMORY.md |
TOML parse; memory/guidance scan; role smoke. |
.claude/agents/level-designer.md |
sonnet | .codex/agents/level-designer.toml |
level-designer |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/live-ops-designer.md |
sonnet | .codex/agents/live-ops-designer.toml |
live-ops-designer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/localization-lead.md |
sonnet | .codex/agents/localization-lead.toml |
localization-lead |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/narrative-director.md |
sonnet | .codex/agents/narrative-director.toml |
narrative-director |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/network-programmer.md |
sonnet | .codex/agents/network-programmer.toml |
network-programmer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/performance-analyst.md |
sonnet | .codex/agents/performance-analyst.toml |
performance-analyst |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/producer.md |
opus | .codex/agents/producer.toml |
producer |
ported | partial | preserve pinned skills sprint-plan, scope-check, estimate, milestone-review; model map gpt-5.5, high |
TOML parse; pinned-skill guidance scan; role smoke. |
.claude/agents/prototyper.md |
sonnet | .codex/agents/prototyper.toml |
prototyper |
ported | partial | upstream has isolation: worktree; make worktree creation explicit |
TOML parse; worktree workflow smoke/manual check. |
.claude/agents/qa-lead.md |
sonnet | .codex/agents/qa-lead.toml |
qa-lead |
ported | partial | preserve pinned skills bug-report, release-checklist |
TOML parse; pinned-skill guidance scan; role smoke. |
.claude/agents/qa-tester.md |
sonnet | .codex/agents/qa-tester.toml |
qa-tester |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/release-manager.md |
sonnet | .codex/agents/release-manager.toml |
release-manager |
ported | partial | preserve pinned skills release-checklist, changelog, patch-notes |
TOML parse; pinned-skill guidance scan; role smoke. |
.claude/agents/security-engineer.md |
sonnet | .codex/agents/security-engineer.toml |
security-engineer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/sound-designer.md |
sonnet | .codex/agents/sound-designer.toml |
sound-designer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/systems-designer.md |
sonnet | .codex/agents/systems-designer.toml |
systems-designer |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/technical-artist.md |
sonnet | .codex/agents/technical-artist.toml |
technical-artist |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/technical-director.md |
opus | .codex/agents/technical-director.toml |
technical-director |
ported | semantic equivalent | model map: gpt-5.5, high; preserve user-memory intent as docs only |
TOML parse; field check; role smoke. |
.claude/agents/tools-programmer.md |
sonnet | .codex/agents/tools-programmer.toml |
tools-programmer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/ue-blueprint-specialist.md |
sonnet | .codex/agents/ue-blueprint-specialist.toml |
ue-blueprint-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/ue-gas-specialist.md |
sonnet | .codex/agents/ue-gas-specialist.toml |
ue-gas-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/ue-replication-specialist.md |
sonnet | .codex/agents/ue-replication-specialist.toml |
ue-replication-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/ue-umg-specialist.md |
sonnet | .codex/agents/ue-umg-specialist.toml |
ue-umg-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/ui-programmer.md |
sonnet | .codex/agents/ui-programmer.toml |
ui-programmer |
ported | semantic equivalent | model map: gpt-5.4, medium |
TOML parse; field check; role smoke. |
.claude/agents/unity-addressables-specialist.md |
sonnet | .codex/agents/unity-addressables-specialist.toml |
unity-addressables-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/unity-dots-specialist.md |
sonnet | .codex/agents/unity-dots-specialist.toml |
unity-dots-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/unity-shader-specialist.md |
sonnet | .codex/agents/unity-shader-specialist.toml |
unity-shader-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/unity-specialist.md |
sonnet | .codex/agents/unity-specialist.toml |
unity-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/unity-ui-specialist.md |
sonnet | .codex/agents/unity-ui-specialist.toml |
unity-ui-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/unreal-specialist.md |
sonnet | .codex/agents/unreal-specialist.toml |
unreal-specialist |
ported | semantic equivalent | engine specialist role | TOML parse; field check; role smoke. |
.claude/agents/ux-designer.md |
sonnet | .codex/agents/ux-designer.toml |
ux-designer |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/world-builder.md |
sonnet | .codex/agents/world-builder.toml |
world-builder |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
.claude/agents/writer.md |
sonnet | .codex/agents/writer.toml |
writer |
ported | semantic equivalent | preserve project-memory intent in instructions | TOML parse; field check; role smoke. |
Common mapping for all rows: upstream behavior is a user-invocable Claude skill or slash-command equivalent; Claude mechanism is .claude/skills/<name>/SKILL.md; Codex mechanism is a repo skill at .agents/skills/<name>/SKILL.md with name: <name>. Claude-only metadata is moved into the body as ported metadata. Task, AskUserQuestion, bare /skill references, and .claude/ paths must be rewritten. Same-name skill collisions must be reported; the port must not silently prefix or rename skills.
| Upstream skill | Model tier | Codex target | Codex skill name | Disposition | Parity | Notes | Verification |
|---|---|---|---|---|---|---|---|
.claude/skills/adopt/SKILL.md |
sonnet | .agents/skills/adopt/SKILL.md |
adopt |
ported | semantic equivalent | Brownfield onboarding workflow | Metadata lint; rewrite scan; smoke as needed. |
.claude/skills/architecture-decision/SKILL.md |
sonnet | .agents/skills/architecture-decision/SKILL.md |
architecture-decision |
ported | semantic equivalent | ADR authoring | Metadata lint; rewrite scan; smoke as needed. |
.claude/skills/architecture-review/SKILL.md |
opus | .agents/skills/architecture-review/SKILL.md |
architecture-review |
ported | partial | High-reasoning skill; delegate to high-tier agents or instruct model choice | Metadata lint; model-tier check; smoke. |
.claude/skills/art-bible/SKILL.md |
sonnet | .agents/skills/art-bible/SKILL.md |
art-bible |
ported | semantic equivalent | Art bible authoring | Metadata lint; rewrite scan. |
.claude/skills/asset-audit/SKILL.md |
sonnet | .agents/skills/asset-audit/SKILL.md |
asset-audit |
ported | semantic equivalent | Asset compliance audit | Metadata lint; rewrite scan. |
.claude/skills/asset-spec/SKILL.md |
sonnet | .agents/skills/asset-spec/SKILL.md |
asset-spec |
ported | semantic equivalent | Asset visual spec generation | Metadata lint; rewrite scan. |
.claude/skills/balance-check/SKILL.md |
sonnet | .agents/skills/balance-check/SKILL.md |
balance-check |
ported | semantic equivalent | Balance analysis | Metadata lint; rewrite scan. |
.claude/skills/brainstorm/SKILL.md |
sonnet | .agents/skills/brainstorm/SKILL.md |
brainstorm |
ported | partial | Structured questions rewritten as numbered options | Metadata lint; AskUserQuestion rewrite scan; smoke. |
.claude/skills/bug-report/SKILL.md |
sonnet | .agents/skills/bug-report/SKILL.md |
bug-report |
ported | semantic equivalent | Bug report workflow | Metadata lint; rewrite scan. |
.claude/skills/bug-triage/SKILL.md |
sonnet | .agents/skills/bug-triage/SKILL.md |
bug-triage |
ported | semantic equivalent | Bug triage workflow | Metadata lint; rewrite scan. |
.claude/skills/changelog/SKILL.md |
haiku | .agents/skills/changelog/SKILL.md |
changelog |
ported | partial | Low-tier model hint is advisory in main-thread skill | Metadata lint; model note scan. |
.claude/skills/code-review/SKILL.md |
sonnet | .agents/skills/code-review/SKILL.md |
code-review |
ported | semantic equivalent | Review workflow | Metadata lint; rewrite scan. |
.claude/skills/consistency-check/SKILL.md |
sonnet | .agents/skills/consistency-check/SKILL.md |
consistency-check |
ported | semantic equivalent | Registry/GDD consistency | Metadata lint; rewrite scan. |
.claude/skills/content-audit/SKILL.md |
sonnet | .agents/skills/content-audit/SKILL.md |
content-audit |
ported | semantic equivalent | Content-count audit | Metadata lint; rewrite scan. |
.claude/skills/create-architecture/SKILL.md |
sonnet | .agents/skills/create-architecture/SKILL.md |
create-architecture |
ported | partial | Structured authoring questions rewritten as numbered options | Metadata lint; smoke. |
.claude/skills/create-control-manifest/SKILL.md |
sonnet | .agents/skills/create-control-manifest/SKILL.md |
create-control-manifest |
ported | semantic equivalent | Control manifest generation | Metadata lint; rewrite scan. |
.claude/skills/create-epics/SKILL.md |
sonnet | .agents/skills/create-epics/SKILL.md |
create-epics |
ported | semantic equivalent | Epic creation | Metadata lint; rewrite scan. |
.claude/skills/create-stories/SKILL.md |
sonnet | .agents/skills/create-stories/SKILL.md |
create-stories |
ported | semantic equivalent | Story breakdown | Metadata lint; rewrite scan. |
.claude/skills/day-one-patch/SKILL.md |
sonnet | .agents/skills/day-one-patch/SKILL.md |
day-one-patch |
ported | semantic equivalent | Launch patch workflow | Metadata lint; rewrite scan. |
.claude/skills/design-review/SKILL.md |
sonnet | .agents/skills/design-review/SKILL.md |
design-review |
ported | semantic equivalent | Design review | Metadata lint; rewrite scan. |
.claude/skills/design-system/SKILL.md |
sonnet | .agents/skills/design-system/SKILL.md |
design-system |
ported | partial | Structured GDD authoring questions rewritten | Metadata lint; smoke. |
.claude/skills/dev-story/SKILL.md |
sonnet | .agents/skills/dev-story/SKILL.md |
dev-story |
ported | semantic equivalent | Story implementation lifecycle | Metadata lint; smoke. |
.claude/skills/estimate/SKILL.md |
sonnet | .agents/skills/estimate/SKILL.md |
estimate |
ported | semantic equivalent | Effort estimate | Metadata lint; rewrite scan. |
.claude/skills/gate-check/SKILL.md |
opus | .agents/skills/gate-check/SKILL.md |
gate-check |
ported | partial | High-reasoning gate; preserve verdict contracts | Metadata lint; model-tier check; smoke. |
.claude/skills/help/SKILL.md |
haiku | .agents/skills/help/SKILL.md |
help |
ported | partial | Low-tier hint advisory | Metadata lint; rewrite scan. |
.claude/skills/hotfix/SKILL.md |
sonnet | .agents/skills/hotfix/SKILL.md |
hotfix |
ported | semantic equivalent | Emergency fix workflow | Metadata lint; rewrite scan. |
.claude/skills/launch-checklist/SKILL.md |
sonnet | .agents/skills/launch-checklist/SKILL.md |
launch-checklist |
ported | semantic equivalent | Launch readiness | Metadata lint; rewrite scan. |
.claude/skills/localize/SKILL.md |
sonnet | .agents/skills/localize/SKILL.md |
localize |
ported | semantic equivalent | Localization workflow | Metadata lint; rewrite scan. |
.claude/skills/map-systems/SKILL.md |
sonnet | .agents/skills/map-systems/SKILL.md |
map-systems |
ported | semantic equivalent | System decomposition | Metadata lint; rewrite scan. |
.claude/skills/milestone-review/SKILL.md |
sonnet | .agents/skills/milestone-review/SKILL.md |
milestone-review |
ported | semantic equivalent | Milestone review | Metadata lint; rewrite scan. |
.claude/skills/onboard/SKILL.md |
haiku | .agents/skills/onboard/SKILL.md |
onboard |
ported | partial | Low-tier hint advisory | Metadata lint; rewrite scan. |
.claude/skills/patch-notes/SKILL.md |
haiku | .agents/skills/patch-notes/SKILL.md |
patch-notes |
ported | partial | Low-tier hint advisory | Metadata lint; rewrite scan. |
.claude/skills/perf-profile/SKILL.md |
sonnet | .agents/skills/perf-profile/SKILL.md |
perf-profile |
ported | semantic equivalent | Performance profiling | Metadata lint; rewrite scan. |
.claude/skills/playtest-report/SKILL.md |
sonnet | .agents/skills/playtest-report/SKILL.md |
playtest-report |
ported | semantic equivalent | Playtest report generation | Metadata lint; rewrite scan. |
.claude/skills/project-stage-detect/SKILL.md |
haiku | .agents/skills/project-stage-detect/SKILL.md |
project-stage-detect |
ported | partial | Low-tier hint advisory; stage detection must stay neutral-path based | Metadata lint; smoke. |
.claude/skills/propagate-design-change/SKILL.md |
sonnet | .agents/skills/propagate-design-change/SKILL.md |
propagate-design-change |
ported | semantic equivalent | Design-change propagation | Metadata lint; rewrite scan. |
.claude/skills/prototype/SKILL.md |
sonnet | .agents/skills/prototype/SKILL.md |
prototype |
ported | partial | Upstream isolation: worktree; make worktree setup explicit |
Metadata lint; worktree smoke/manual check. |
.claude/skills/qa-plan/SKILL.md |
sonnet | .agents/skills/qa-plan/SKILL.md |
qa-plan |
ported | semantic equivalent | QA plan | Metadata lint; rewrite scan. |
.claude/skills/quick-design/SKILL.md |
sonnet | .agents/skills/quick-design/SKILL.md |
quick-design |
ported | partial | Structured options rewritten | Metadata lint; rewrite scan. |
.claude/skills/regression-suite/SKILL.md |
sonnet | .agents/skills/regression-suite/SKILL.md |
regression-suite |
ported | semantic equivalent | Regression plan | Metadata lint; rewrite scan. |
.claude/skills/release-checklist/SKILL.md |
sonnet | .agents/skills/release-checklist/SKILL.md |
release-checklist |
ported | semantic equivalent | Release readiness | Metadata lint; rewrite scan. |
.claude/skills/retrospective/SKILL.md |
sonnet | .agents/skills/retrospective/SKILL.md |
retrospective |
ported | semantic equivalent | Retro workflow | Metadata lint; rewrite scan. |
.claude/skills/reverse-document/SKILL.md |
sonnet | .agents/skills/reverse-document/SKILL.md |
reverse-document |
ported | semantic equivalent | Reverse documentation | Metadata lint; rewrite scan. |
.claude/skills/review-all-gdds/SKILL.md |
opus | .agents/skills/review-all-gdds/SKILL.md |
review-all-gdds |
ported | partial | High-reasoning review; preserve multi-GDD gate semantics | Metadata lint; model-tier check; smoke. |
.claude/skills/scope-check/SKILL.md |
haiku | .agents/skills/scope-check/SKILL.md |
scope-check |
ported | partial | Low-tier hint advisory | Metadata lint; rewrite scan. |
.claude/skills/security-audit/SKILL.md |
sonnet | .agents/skills/security-audit/SKILL.md |
security-audit |
ported | semantic equivalent | Security audit | Metadata lint; rewrite scan. |
.claude/skills/setup-engine/SKILL.md |
sonnet | .agents/skills/setup-engine/SKILL.md |
setup-engine |
ported | partial | Structured engine questions rewritten | Metadata lint; smoke. |
.claude/skills/skill-improve/SKILL.md |
sonnet | .agents/skills/skill-improve/SKILL.md |
skill-improve |
ported | semantic equivalent | Skill improvement loop | Metadata lint; rewrite scan. |
.claude/skills/skill-test/SKILL.md |
sonnet | .agents/skills/skill-test/SKILL.md |
skill-test |
ported | semantic equivalent | Upstream skill, not new | Metadata lint; testing-framework reference scan. |
.claude/skills/smoke-check/SKILL.md |
sonnet | .agents/skills/smoke-check/SKILL.md |
smoke-check |
ported | semantic equivalent | Smoke gate | Metadata lint; smoke. |
.claude/skills/soak-test/SKILL.md |
sonnet | .agents/skills/soak-test/SKILL.md |
soak-test |
ported | semantic equivalent | Soak protocol | Metadata lint; rewrite scan. |
.claude/skills/sprint-plan/SKILL.md |
sonnet | .agents/skills/sprint-plan/SKILL.md |
sprint-plan |
ported | semantic equivalent | Sprint planning | Metadata lint; smoke. |
.claude/skills/sprint-status/SKILL.md |
haiku | .agents/skills/sprint-status/SKILL.md |
sprint-status |
ported | partial | Low-tier hint advisory | Metadata lint; smoke. |
.claude/skills/start/SKILL.md |
sonnet | .agents/skills/start/SKILL.md |
start |
ported | semantic equivalent | First-time onboarding entry point | Metadata lint; end-to-end onboarding smoke. |
.claude/skills/story-done/SKILL.md |
sonnet | .agents/skills/story-done/SKILL.md |
story-done |
ported | semantic equivalent | End-of-story review | Metadata lint; smoke. |
.claude/skills/story-readiness/SKILL.md |
sonnet | .agents/skills/story-readiness/SKILL.md |
story-readiness |
ported | semantic equivalent | Story readiness gate | Metadata lint; smoke. |
.claude/skills/team-audio/SKILL.md |
sonnet | .agents/skills/team-audio/SKILL.md |
team-audio |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/team-combat/SKILL.md |
sonnet | .agents/skills/team-combat/SKILL.md |
team-combat |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/team-level/SKILL.md |
sonnet | .agents/skills/team-level/SKILL.md |
team-level |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/team-live-ops/SKILL.md |
sonnet | .agents/skills/team-live-ops/SKILL.md |
team-live-ops |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/team-narrative/SKILL.md |
sonnet | .agents/skills/team-narrative/SKILL.md |
team-narrative |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/team-polish/SKILL.md |
sonnet | .agents/skills/team-polish/SKILL.md |
team-polish |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/team-qa/SKILL.md |
sonnet | .agents/skills/team-qa/SKILL.md |
team-qa |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/team-release/SKILL.md |
sonnet | .agents/skills/team-release/SKILL.md |
team-release |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/team-ui/SKILL.md |
sonnet | .agents/skills/team-ui/SKILL.md |
team-ui |
ported | semantic equivalent | Team orchestration | Metadata lint; delegation reference check. |
.claude/skills/tech-debt/SKILL.md |
sonnet | .agents/skills/tech-debt/SKILL.md |
tech-debt |
ported | semantic equivalent | Tech debt tracking | Metadata lint; rewrite scan. |
.claude/skills/test-evidence-review/SKILL.md |
sonnet | .agents/skills/test-evidence-review/SKILL.md |
test-evidence-review |
ported | semantic equivalent | Test evidence review | Metadata lint; rewrite scan. |
.claude/skills/test-flakiness/SKILL.md |
sonnet | .agents/skills/test-flakiness/SKILL.md |
test-flakiness |
ported | semantic equivalent | Flakiness triage | Metadata lint; rewrite scan. |
.claude/skills/test-helpers/SKILL.md |
sonnet | .agents/skills/test-helpers/SKILL.md |
test-helpers |
ported | semantic equivalent | Test helper generation | Metadata lint; rewrite scan. |
.claude/skills/test-setup/SKILL.md |
sonnet | .agents/skills/test-setup/SKILL.md |
test-setup |
ported | semantic equivalent | Test/CI scaffold | Metadata lint; rewrite scan. |
.claude/skills/ux-design/SKILL.md |
sonnet | .agents/skills/ux-design/SKILL.md |
ux-design |
ported | partial | Structured UX authoring questions rewritten | Metadata lint; smoke. |
.claude/skills/ux-review/SKILL.md |
sonnet | .agents/skills/ux-review/SKILL.md |
ux-review |
ported | semantic equivalent | UX review | Metadata lint; rewrite scan. |
.claude/skills/vertical-slice/SKILL.md |
sonnet | .agents/skills/vertical-slice/SKILL.md |
vertical-slice |
ported | partial | Upstream isolation: worktree; framework spec missing upstream |
Metadata lint; worktree smoke/manual check; missing spec check. |
| New Codex status skill | n/a | .agents/skills/studio-status/SKILL.md |
studio-status |
ported | new Codex-native support | Complements native tui.status_line by rendering the game-stage/review/session breadcrumb on demand |
Metadata lint; reads shared state. |
| New Codex continuity skill | n/a | .agents/skills/studio-next/SKILL.md |
studio-next |
ported | new Codex-native support | Reads handoff, active session, sprint, stage, workflow, and slice state to recommend the single best next action after discrete work | Metadata lint; read-only state routing smoke. |
Codex has global/thread Memories under ~/.codex/memories, but no documented direct equivalent of Claude's per-agent memory: user or memory: project binding. The port therefore preserves memory scope as repo-local agent memory contracts and explicit custom-agent instructions. The installer must not write global Codex memories.
| Upstream memory source | Scope | Codex target | Disposition | Parity | Implementation notes | Verification |
|---|---|---|---|---|---|---|
.claude/agents/creative-director.md memory: user |
user | .codex/agent-memory/creative-director/MEMORY.md plus .codex/agents/creative-director.toml binding |
replaced | semantic equivalent | Generated memory contract for durable creative taste, vision preferences, pillar rulings, and recurring user-level creative constraints. Label as generated; do not invent historical upstream memories. | Agent TOML includes Ported Claude memory scope: user; memory file exists; no ~/.codex/memories writes. |
.claude/agents/technical-director.md memory: user |
user | .codex/agent-memory/technical-director/MEMORY.md plus .codex/agents/technical-director.toml binding |
replaced | semantic equivalent | Generated memory contract for durable engine/tooling preferences, architecture principles, platform constraints, and recurring technical standards. | TOML binding scan; memory file exists. |
.claude/agents/producer.md memory: user |
user | .codex/agent-memory/producer/MEMORY.md plus .codex/agents/producer.toml binding |
replaced | semantic equivalent | Generated memory contract for planning style, scope tolerance, schedule/risk preferences, and recurring production constraints. | TOML binding scan; memory file exists. |
.claude/agents/art-director.md memory: project |
project | .codex/agent-memory/art-director/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for visual identity, art bible decisions, asset constraints, and art pipeline preferences. | TOML binding scan; memory file exists. |
.claude/agents/audio-director.md memory: project |
project | .codex/agent-memory/audio-director/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for audio identity, mix rules, middleware, and sound bible decisions. | TOML binding scan; memory file exists. |
.claude/agents/economy-designer.md memory: project |
project | .codex/agent-memory/economy-designer/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for balance, economy formulas, progression rules, and monetization constraints. | TOML binding scan; memory file exists. |
.claude/agents/game-designer.md memory: project |
project | .codex/agent-memory/game-designer/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for mechanics, GDD conventions, system boundaries, and gameplay pillars. | TOML binding scan; memory file exists. |
.claude/agents/lead-programmer.md plus .claude/agent-memory/lead-programmer/MEMORY.md |
project | .codex/agent-memory/lead-programmer/MEMORY.md plus .codex/agents/lead-programmer.toml binding |
ported | semantic equivalent | Rewrite the explicit upstream memory file for Codex paths and metadata expectations. | Content rewrite scan; TOML binding scan. |
.claude/agents/level-designer.md memory: project |
project | .codex/agent-memory/level-designer/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for level layout conventions, pacing, encounter grammar, and spatial constraints. | TOML binding scan; memory file exists. |
.claude/agents/localization-lead.md memory: project |
project | .codex/agent-memory/localization-lead/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for locale policy, string pipeline, glossary, and translation memory conventions. | TOML binding scan; memory file exists. |
.claude/agents/narrative-director.md memory: project |
project | .codex/agent-memory/narrative-director/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for canon, voice, story constraints, and narrative bible decisions. | TOML binding scan; memory file exists. |
.claude/agents/performance-analyst.md memory: project |
project | .codex/agent-memory/performance-analyst/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for frame budgets, profiling baselines, memory ceilings, and known bottlenecks. | TOML binding scan; memory file exists. |
.claude/agents/qa-lead.md memory: project |
project | .codex/agent-memory/qa-lead/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for quality gates, test strategy, defect policy, and release criteria. | TOML binding scan; memory file exists. |
.claude/agents/systems-designer.md memory: project |
project | .codex/agent-memory/systems-designer/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for system taxonomy, dependencies, balance constraints, and design consistency rules. | TOML binding scan; memory file exists. |
.claude/agents/ux-designer.md memory: project |
project | .codex/agent-memory/ux-designer/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for UX patterns, accessibility choices, HUD conventions, and interaction decisions. | TOML binding scan; memory file exists. |
.claude/agents/world-builder.md memory: project |
project | .codex/agent-memory/world-builder/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for world rules, factions, locations, lore constraints, and setting continuity. | TOML binding scan; memory file exists. |
.claude/agents/writer.md memory: project |
project | .codex/agent-memory/writer/MEMORY.md plus TOML binding |
replaced | semantic equivalent | Generated project memory contract for character voice, dialogue style, terminology, and narrative continuity. | TOML binding scan; memory file exists. |
All 40 upstream files under .claude/docs/templates/** are copied to .codex/docs/templates/** with the same relative path. Parity is exact unless the template text contains Claude command references, in which case those references are rewritten while preserving the artifact schema.
| Upstream template | Target | Disposition | Verification |
|---|---|---|---|
accessibility-requirements.md |
.codex/docs/templates/accessibility-requirements.md |
ported | Manifest diff; content rewrite scan. |
architecture-decision-record.md |
.codex/docs/templates/architecture-decision-record.md |
ported | Manifest diff; content rewrite scan. |
architecture-doc-from-code.md |
.codex/docs/templates/architecture-doc-from-code.md |
ported | Manifest diff; content rewrite scan. |
architecture-traceability.md |
.codex/docs/templates/architecture-traceability.md |
ported | Manifest diff; content rewrite scan. |
art-bible.md |
.codex/docs/templates/art-bible.md |
ported | Manifest diff; content rewrite scan. |
changelog-template.md |
.codex/docs/templates/changelog-template.md |
ported | Manifest diff; content rewrite scan. |
collaborative-protocols/design-agent-protocol.md |
.codex/docs/templates/collaborative-protocols/design-agent-protocol.md |
ported | Manifest diff; content rewrite scan. |
collaborative-protocols/implementation-agent-protocol.md |
.codex/docs/templates/collaborative-protocols/implementation-agent-protocol.md |
ported | Manifest diff; content rewrite scan. |
collaborative-protocols/leadership-agent-protocol.md |
.codex/docs/templates/collaborative-protocols/leadership-agent-protocol.md |
ported | Manifest diff; content rewrite scan. |
concept-doc-from-prototype.md |
.codex/docs/templates/concept-doc-from-prototype.md |
ported | Manifest diff; content rewrite scan. |
design-doc-from-implementation.md |
.codex/docs/templates/design-doc-from-implementation.md |
ported | Manifest diff; content rewrite scan. |
difficulty-curve.md |
.codex/docs/templates/difficulty-curve.md |
ported | Manifest diff; content rewrite scan. |
economy-model.md |
.codex/docs/templates/economy-model.md |
ported | Manifest diff; content rewrite scan. |
faction-design.md |
.codex/docs/templates/faction-design.md |
ported | Manifest diff; content rewrite scan. |
game-concept.md |
.codex/docs/templates/game-concept.md |
ported | Manifest diff; content rewrite scan. |
game-design-document.md |
.codex/docs/templates/game-design-document.md |
ported | Manifest diff; content rewrite scan. |
game-pillars.md |
.codex/docs/templates/game-pillars.md |
ported | Manifest diff; content rewrite scan. |
hud-design.md |
.codex/docs/templates/hud-design.md |
ported | Manifest diff; content rewrite scan. |
incident-response.md |
.codex/docs/templates/incident-response.md |
ported | Manifest diff; content rewrite scan. |
interaction-pattern-library.md |
.codex/docs/templates/interaction-pattern-library.md |
ported | Manifest diff; content rewrite scan. |
level-design-document.md |
.codex/docs/templates/level-design-document.md |
ported | Manifest diff; content rewrite scan. |
milestone-definition.md |
.codex/docs/templates/milestone-definition.md |
ported | Manifest diff; content rewrite scan. |
narrative-character-sheet.md |
.codex/docs/templates/narrative-character-sheet.md |
ported | Manifest diff; content rewrite scan. |
pitch-document.md |
.codex/docs/templates/pitch-document.md |
ported | Manifest diff; content rewrite scan. |
player-journey.md |
.codex/docs/templates/player-journey.md |
ported | Manifest diff; content rewrite scan. |
post-mortem.md |
.codex/docs/templates/post-mortem.md |
ported | Manifest diff; content rewrite scan. |
project-stage-report.md |
.codex/docs/templates/project-stage-report.md |
ported | Manifest diff; content rewrite scan. |
prototype-report.md |
.codex/docs/templates/prototype-report.md |
ported | Manifest diff; content rewrite scan. |
release-checklist-template.md |
.codex/docs/templates/release-checklist-template.md |
ported | Manifest diff; content rewrite scan. |
release-notes.md |
.codex/docs/templates/release-notes.md |
ported | Manifest diff; content rewrite scan. |
risk-register-entry.md |
.codex/docs/templates/risk-register-entry.md |
ported | Manifest diff; content rewrite scan. |
skill-test-spec.md |
.codex/docs/templates/skill-test-spec.md |
ported | Manifest diff; content rewrite scan. |
sound-bible.md |
.codex/docs/templates/sound-bible.md |
ported | Manifest diff; content rewrite scan. |
sprint-plan.md |
.codex/docs/templates/sprint-plan.md |
ported | Manifest diff; content rewrite scan. |
systems-index.md |
.codex/docs/templates/systems-index.md |
ported | Manifest diff; content rewrite scan. |
technical-design-document.md |
.codex/docs/templates/technical-design-document.md |
ported | Manifest diff; content rewrite scan. |
test-evidence.md |
.codex/docs/templates/test-evidence.md |
ported | Manifest diff; content rewrite scan. |
test-plan.md |
.codex/docs/templates/test-plan.md |
ported | Manifest diff; content rewrite scan. |
ux-spec.md |
.codex/docs/templates/ux-spec.md |
ported | Manifest diff; content rewrite scan. |
vertical-slice-report.md |
.codex/docs/templates/vertical-slice-report.md |
ported | Manifest diff; content rewrite scan. |
| Upstream files | Count | Target | Disposition | Parity | Notes | Verification |
|---|---|---|---|---|---|---|
.claude/docs/*.md, .claude/docs/hooks-reference/*.md, .claude/docs/workflow-catalog.yaml excluding templates |
23 | .codex/docs/** with rewritten runtime paths |
ported | semantic equivalent | Includes agent-coordination-map.md, agent-roster.md, coding-standards.md, context-management.md, coordination-rules.md, director-gates.md, directory-structure.md, hooks-reference*.md, quick-start.md, review-workflow.md, rules-reference.md, setup-requirements.md, skills-reference.md, technical-preferences.md, workflow-catalog.yaml, plus local template docs. |
Manifest path diff; YAML parse; no runtime .claude refs except provenance allowlist. |
README.md, UPGRADING.md, CONTRIBUTING.md, SECURITY.md |
4 | .codex/docs/README.md, .codex/docs/MIGRATION.md, .codex/docs/CONTRIBUTING.md, .codex/docs/SECURITY.md |
ported | semantic equivalent | Rewrite Claude invocation/install instructions to Codex; preserve upstream attribution. | Docs grep; link check. |
LICENSE |
1 | LICENSE and ATTRIBUTION.md |
shared/ported | exact | Preserve MIT license and source attribution. | License text present; attribution present. |
.gitignore |
1 | Optional documented additions only | not applicable | not applicable | Do not overwrite a target game's .gitignore; installer may suggest Codex log/backups ignores. |
Install fixture asserts .gitignore unchanged unless user opts in. |
.github/CODEOWNERS, .github/FUNDING.yml, .github/PULL_REQUEST_TEMPLATE.md, issue templates |
5 | docs/github/** or optional distribution files |
ported | semantic equivalent | Not runtime Codex assets; useful for source distribution and contribution docs. | Manifest coverage; no install overwrite by default. |
docs/COLLABORATIVE-DESIGN-PRINCIPLE.md, docs/WORKFLOW-GUIDE.md, docs/architecture/tr-registry.yaml, docs/registry/architecture.yaml |
4 | same neutral paths under docs/** |
shared | exact | These are project docs/data, not Claude runtime files. | Path existence; YAML parse. |
docs/engine-reference/** |
46 | same neutral paths | shared | exact | Engine references remain shared for Claude and Codex workflows. | Count 46; path diff. |
docs/examples/** |
11 | docs/examples/** or same neutral paths |
ported | semantic equivalent | Rewrite command examples to upstream Codex skill names such as start and prototype. |
Count 11; grep for bare Claude slash commands. |
Target root: CCGS Skill Testing Framework/**.
| Upstream testing asset | Count | Target | Disposition | Parity | Notes | Verification |
|---|---|---|---|---|---|---|
CCGS Skill Testing Framework/README.md, catalog.yaml, quality-rubric.md |
3 | same relative names under target root | ported | semantic equivalent | Rewrite Claude path/runtime references only. | Count/path diff; YAML parse. |
CCGS Skill Testing Framework/CLAUDE.md |
1 | CCGS Skill Testing Framework/AGENTS.md or README section |
replaced | semantic equivalent | Framework guidance becomes Codex instructions/docs. | No .claude dependency. |
CCGS Skill Testing Framework/agents/**/<agent>.md |
49 | same relative tree under target root | ported | semantic equivalent | Agent spec names match all 49 upstream agents. | Agent-spec vs agent manifest diff. |
CCGS Skill Testing Framework/skills/**/<skill>.md |
72 | same relative tree under target root | ported | semantic equivalent | Specs exist for all upstream skills except vertical-slice. |
Skill-spec vs 73-skill manifest diff; inherited missing vertical-slice remains historical evidence. |
Missing upstream vertical-slice skill spec |
1 expected spec absent upstream | CCGS Skill Testing Framework/skills/utility/vertical-slice.md |
ported as Codex-added coverage | partial | Upstream gap, not a Codex conversion miss. This port adds a clearly labeled Codex spec covering PROCEED/PIVOT/KILL expectations. | Validation includes the Codex-added spec while upstream inventory remains historical evidence of the inherited gap. |
CCGS Skill Testing Framework/templates/agent-test-spec.md, skill-test-spec.md |
2 | same relative names under target root | ported | semantic equivalent | Rewrite command examples to Codex naming. | Count/path diff. |
| Behavior/component | Upstream behavior | Claude mechanism | Codex target mechanism | Target | Disposition | Parity | Verification |
|---|---|---|---|---|---|---|---|
| Phase/stage state | Tracks project lifecycle phase | Shared files read by Claude skills/hooks/statusline | Same neutral files read by Codex skills/hooks | production/stage.txt, production/review-mode.txt, production/session-state/active.md |
shared | exact | Byte-level fixture; no Codex namespace fork. |
| Review gates | Gate readiness and director verdicts | Skills plus director docs | gate-check, review-all-gdds, high-tier agents |
Skills/agents plus docs/director-gates.md |
ported | semantic equivalent | Smoke gate workflow; verdict token check. |
| Delegation model | Team/role routing | Claude Task to named agents |
Codex custom agents with upstream role names | .codex/agents/*.toml plus skill rewrites |
ported | semantic equivalent | Role reference resolver; sample subagent smoke. |
| Structured choices | Guided option questions | Claude AskUserQuestion |
Numbered natural-language choices; optional MCP later | Skill bodies | replaced | partial | Grep AskUserQuestion; interactive workflow review. |
| Slash-command workflows | /start, /brainstorm, etc. |
Claude slash commands via skills | Codex skills named start, brainstorm, etc. |
.agents/skills/<skill>/SKILL.md |
replaced | semantic equivalent | Grep for bare custom slash commands; skill selector smoke plus collision audit. |
| Install/upgrade | Upstream is a Claude template, not coexistence installer | Manual clone/template | Idempotent Codex installer/uninstaller | .codex/install.sh, uninstall.sh, audit.sh, manifests |
new Codex-native support | new support | Fixture install/uninstall; hash .claude/**, CLAUDE.md, shared state. |
| Optional plugin package | Not upstream behavior | none | Optional later package | optional .codex-plugin/plugin.json |
not applicable for base port | not applicable | Base validation must not depend on plugin cache/marketplace. |
| MCP | Not upstream behavior | none | Optional only for elicitation or external services | none by default | not applicable | not applicable | No MCP config unless a future phase adds a documented rationale. |
Phase 4 must turn this matrix into checks for:
- 417 upstream files categorized by manifest.
- 49 generated custom agent TOML files.
- 73 ported skills plus 4 new Codex-native support skills:
studio-status,studio-next,handoff, andresume-from-handoff. - 11 ported hook scripts plus
notify.shreplaced by native notification setup documentation. - 15 root-routed Codex path-rule documents, including the 11 upstream rule mappings.
- 40 copied templates.
- 127 upstream testing-framework assets plus 1 Codex-added skill spec for
vertical-slice. - Zero runtime dependency on
.claude/,CLAUDE.md, or bare upstream slash commands.