@jstn-sdk/rcs 0.1.7 → 0.1.8

package/skills/hud/SKILL.md

@@ -100,6 +100,3 @@ If the TUI statusline is not showing:
 If `rcs hud` shows "No active modes":
 - This is expected when no workflows are running
 - Start a workflow (forge, autopilot, etc.) and check again
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/note/SKILL.md

@@ -64,6 +64,3 @@ Notepad content is automatically loaded on session start:
 - Working Memory: Loaded if recent entries exist
 
 This helps survive conversation compaction without losing critical context.
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/pipeline/SKILL.md

@@ -88,6 +88,3 @@ const result = await runPipeline(config);
 - **team**: Pipeline delegates execution to team mode (Codex CLI workers)
 - **forge**: Pipeline delegates verification to forge (configurable iterations)
 - **blueprint**: Pipeline's first stage runs consensus planning
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/plan/SKILL.md

@@ -279,6 +279,3 @@ Before asking any interview question, classify it:
 
 The separate `/planner` and `/review` skills have been merged into `$plan`. Consensus planning remains available through `$plan --consensus`, while `$blueprint` is the canonical creator-facing planning workflow when you want the full Roblox-native design lane.
 </Advanced>
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/rcs-setup/SKILL.md

@@ -137,6 +137,3 @@ node bin/rcs.js doctor
 
 - If AGENTS.md was not overwritten during `--force`, stop active RCS session and rerun setup.
 - If AGENTS.md was not merged during `--merge-agents`, stop active RCS session and rerun setup.
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/review/SKILL.md

@@ -40,6 +40,3 @@ The review workflow:
 - Approval must cite concrete evidence, not author claims.
 
 Follow the Plan skill's full documentation for review mode details.
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/security-review/SKILL.md

@@ -179,6 +179,3 @@ OVERALL: [POSTURE] — ship / fix-first / block live
 **Good:** The user says `continue` after you already listed the next remote to trace; keep tracing until the chain is complete.
 
 **Bad:** The user says `continue` and you restart from “what is Roblox” instead of finishing the active exploit hypothesis.
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/skill/SKILL.md

@@ -11,6 +11,7 @@ artifact-type: "skill"
 # Skill Management CLI
 
 Meta-skill for managing Roblox Creator Skills workflow skills via CLI-like commands.
+Default examples should stay grounded in Roblox creator workflows rather than generic software helpers.
 
 ## Subcommands
 
@@ -28,13 +29,13 @@ Show all local skills organized by scope.
 USER SKILLS (~/.codex/skills/):
 | Name              | Triggers           | Quality | Usage | Scope |
 |-------------------|--------------------|---------|-------|-------|
-| error-handler     | fix, error         | 95%     | 42    | user  |
-| remote-audit      | remote, payload    | 88%     | 23    | user  |
+| remote-audit      | remote, payload    | 95%     | 42    | user  |
+| gui-state-check   | gui, hud, state    | 88%     | 23    | user  |
 
 PROJECT SKILLS (.codex/skills/):
 | Name              | Triggers           | Quality | Usage | Scope   |
 |-------------------|--------------------|---------|-------|---------|
-| test-runner       | test, run          | 92%     | 15    | project |
+| datastore-guard   | datastore, profile | 92%     | 15    | project |
 ```
 
 **Fallback:** If quality/usage stats not available, show "N/A"
@@ -101,16 +102,16 @@ argument-hint: "<args>"
 
 **Example:**
 ```
-User: /skill add custom-logger
-Assistant: Creating new skill 'custom-logger'...
+User: /skill add remote-audit-helper
+Assistant: Creating new skill 'remote-audit-helper'...
 
-Description: Enhanced logging with structured output
-Triggers (comma-separated): log, logger, logging
-Argument hint (optional): <level> [message]
+Description: Audit RemoteEvent payload validation and server authority checks
+Triggers (comma-separated): remote, payload, validation
+Argument hint (optional): <remote-name> [surface]
 Scope (user/project): user
 
-✓ Created skill at ~/.codex/skills/custom-logger/SKILL.md
-→ Edit with: /skill edit custom-logger
+✓ Created skill at ~/.codex/skills/remote-audit-helper/SKILL.md
+→ Edit with: /skill edit remote-audit-helper
 ```
 
 ---
@@ -136,15 +137,15 @@ Remove a skill by name.
 
 **Example:**
 ```
-User: /skill remove old-logger
-Assistant: Found skill 'old-logger' in user scope:
-  Description: Legacy logging utility
+User: /skill remove old-trade-audit
+Assistant: Found skill 'old-trade-audit' in user scope:
+  Description: Legacy trading remote audit helper
   Scope: user (~/.codex/skills/)
 
-Delete 'old-logger' skill? (yes/no)
+Delete 'old-trade-audit' skill? (yes/no)
 
 User: yes
-Assistant: ✓ Removed skill 'old-logger' from user scope
+Assistant: ✓ Removed skill 'old-trade-audit' from user scope
 ```
 
 ---
@@ -158,10 +159,10 @@ Edit an existing skill interactively.
 2. **Read current content** via Read tool
 3. **Display current values:**
    ```
-   Current skill 'custom-logger':
-   - Description: Enhanced logging with structured output
-   - Triggers: log, logger, logging
-   - Argument hint: <level> [message]
+   Current skill 'remote-audit-helper':
+   - Description: Audit RemoteEvent payload validation and server authority checks
+   - Triggers: remote, payload, validation
+   - Argument hint: <remote-name> [surface]
    - Scope: user
    ```
 4. **Ask what to change:**
@@ -180,19 +181,19 @@ Edit an existing skill interactively.
 
 **Example:**
 ```
-User: /skill edit custom-logger
-Assistant: Current skill 'custom-logger':
-  - Description: Enhanced logging with structured output
-  - Triggers: log, logger, logging
-  - Argument hint: <level> [message]
+User: /skill edit remote-audit-helper
+Assistant: Current skill 'remote-audit-helper':
+  - Description: Audit RemoteEvent payload validation and server authority checks
+  - Triggers: remote, payload, validation
+  - Argument hint: <remote-name> [surface]
 
 What would you like to edit? (description/triggers/argument-hint/content/rename/cancel)
 
 User: triggers
-Assistant: Current triggers: log, logger, logging
-New triggers (comma-separated): log, logger, logging, trace
+Assistant: Current triggers: remote, payload, validation
+New triggers (comma-separated): remote, payload, validation, server-authority
 
-✓ Updated triggers for 'custom-logger'
+✓ Updated triggers for 'remote-audit-helper'
 ```
 
 ---
@@ -211,19 +212,19 @@ Search skills by content, triggers, name, or description.
 3. **Display matches** with context:
 
 ```
-Found 3 skills matching "typescript error":
+Found 3 skills matching "remote payload":
 
-1. typescript-fixer (user)
-   Description: Fix common TypeScript errors
-   Match: "typescript error handling patterns"
+1. remote-audit (user)
+   Description: Review RemoteEvent / RemoteFunction payload contracts
+   Match: "remote payload validation"
 
-2. error-handler (user)
-   Description: Generic error handling utilities
-   Match: "Supports TypeScript and JavaScript errors"
+2. trade-safety (user)
+   Description: Trading safety checks for inventory mutation
+   Match: "payload ownership and duplicate-request validation"
 
-3. lint-fix (project)
-   Description: Auto-fix linting errors
-   Match: "TypeScript ESLint error resolution"
+3. datastore-guard (project)
+   Description: Profile and save-path safety helpers
+   Match: "remote requests that mutate persistent state"
 ```
 
 **Ranking:** Prioritize matches in name/triggers over content matches
@@ -837,6 +838,3 @@ What would you like to do?
 - `/skill stats` - Show usage statistics across all skills
 - `/skill validate` - Check all skills for format errors
 - `/skill template <type>` - Create from predefined templates
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/swarm/SKILL.md

@@ -9,7 +9,7 @@ artifact-type: "skill"
 
 # Swarm (Compatibility Facade)
 
-Swarm is a compatibility alias for the `/team` skill. Public creator language should prefer `$crew` or `$team`; all swarm invocations are routed to the Team skill's staged pipeline.
+Swarm is a historical compatibility alias for the `/team` skill. Active creator language should prefer `$crew` or `$team`; all swarm invocations are routed to the Team skill's staged pipeline.
 
 ## Usage
 
@@ -27,6 +27,3 @@ This skill is identical to `/team`. Invoke the Team skill with the same argument
 ```
 
 Follow the Team skill's full documentation for staged pipeline, agent routing, and coordination semantics.
-surface-class: "historical"
-domain: "creator-runtime"
-audience: "internal"

package/skills/tdd/SKILL.md

@@ -108,6 +108,3 @@ Use `mcp__x__ask_codex` with `agent_role: "tdd-guide"`.
 If ToolSearch finds no MCP tools, fall back to the `test-engineer` agent.
 
 **Remember:** The discipline IS the value. Shortcuts destroy the benefit.
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/team/SKILL.md

@@ -511,6 +511,3 @@ Two cleanup paths exist and must not be confused:
 **Good:** The user changes only the output shape or downstream delivery step (for example `make a PR`). Preserve earlier non-conflicting workflow constraints and apply the update locally.
 
 **Bad:** The user says `continue`, and the workflow restarts discovery or stops before the missing verification/evidence is gathered.
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/trace/SKILL.md

@@ -35,6 +35,3 @@ Present the timeline first, then the summary. Highlight:
 - **Mode transitions** (how execution modes changed)
 - **Bottlenecks** (slow tools or agents)
 - **Flow patterns** (keyword -> skill -> agent chains)
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/ultraqa/SKILL.md

@@ -9,9 +9,9 @@ artifact-type: "skill"
 
 # UltraQA Skill
 
-[ULTRAQA ACTIVATED - AUTONOMOUS QA CYCLING]
+Use `$ultraqa` when you need a strict QA loop for Roblox Studio, Luau, creator-runtime, or plugin changes and the first verification pass is not enough.
 
-## Overview
+This skill is strongest when the work needs repeatable test -> diagnose -> fix -> retest cycling until the changed behavior is actually proven.
 
 ## GPT-5.5 Guidance Alignment
 
@@ -30,7 +30,7 @@ Parse the goal from arguments. Supported formats:
 | `/ultraqa --tests` | tests | All test suites pass |
 | `/ultraqa --build` | build | Build succeeds with exit 0 |
 | `/ultraqa --lint` | lint | No lint errors |
-| `/ultraqa --typecheck` | typecheck | No TypeScript errors |
+| `/ultraqa --typecheck` | typecheck | No TypeScript errors in the RCS runtime/tooling layer |
 | `/ultraqa --custom "pattern"` | custom | Custom success pattern in output |
 
 If no structured goal provided, interpret the argument as a custom goal.
@@ -49,7 +49,7 @@ If no structured goal provided, interpret the argument as a custom goal.
      ```
      delegate(role="test-engineer", tier="STANDARD", task="TEST:
      Goal: [describe what to verify]
-     Service: [how to start]
+     Surface: [Roblox Studio, Luau module, plugin widget, or RCS runtime]
      Test cases: [specific scenarios to verify]")
      ```
 
@@ -89,11 +89,11 @@ If no structured goal provided, interpret the argument as a custom goal.
 Output progress each cycle:
 ```
 [ULTRAQA Cycle 1/5] Running tests...
-[ULTRAQA Cycle 1/5] FAILED - 3 tests failing
+[ULTRAQA Cycle 1/5] FAILED - 3 checks failing
 [ULTRAQA Cycle 1/5] Architect diagnosing...
-[ULTRAQA Cycle 1/5] Fixing: auth.test.ts - missing mock
+[ULTRAQA Cycle 1/5] Fixing: TradeRequest.server.luau validation path or matching runtime harness
 [ULTRAQA Cycle 2/5] Running tests...
-[ULTRAQA Cycle 2/5] PASSED - All 47 tests pass
+[ULTRAQA Cycle 2/5] PASSED - verification target is green
 [ULTRAQA COMPLETE] Goal met after 2 cycles
 ```
 
@@ -145,6 +145,3 @@ Use MCP state cleanup rather than deleting files directly.
 ---
 
 Begin ULTRAQA cycling now. Parse the goal and start cycle 1.
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/ultrawork/SKILL.md

@@ -177,6 +177,3 @@ ecomode (token efficiency)
 
 Ultrawork is the parallelism and execution-discipline layer. Forge adds persistence, architect verification, deslop, and retry-until-done behavior. Autoforge adds the broader autonomous lifecycle pipeline. Ecomode adjusts ultrawork's model routing to favor cheaper models.
 </Advanced>
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/visual-forge/SKILL.md

@@ -9,7 +9,7 @@ artifact-type: "skill"
 
 # Visual Forge Skill
 
-Legacy compatibility surface. Public creator language should prefer **Visual Forge** concepts and `$forge` as the persistent implementation owner.
+Visual Forge is the visual-delivery lane for Roblox creator UI, HUD, and plugin work. `$forge` remains the persistent implementation owner when the work turns into a code-bearing completion loop.
 
 Use this skill when the user wants Codex to build or restyle Roblox creator UI through a compatibility visual-delivery loop: an approved generated reference, static reference, or captured inspiration baseline becomes the target, Forge implements, and Visual Verdict drives measured iteration rather than subjective description alone.
 
@@ -157,6 +157,3 @@ Do not make major design pivots unless explicitly requested."
 ```
 
 Task: {{ARGUMENTS}}
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/visual-verdict/SKILL.md

@@ -78,6 +78,3 @@ When mismatch diagnosis is hard:
 }
 ```
 </Example>
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/web-clone/SKILL.md

@@ -26,6 +26,3 @@ If this skill is selected by older routing, stop the standalone web-clone pipeli
 ## Rationale
 
 The URL extraction, visual iteration, and implementation verification responsibilities are now part of Visual Forge's broader visual-delivery workflow. Keeping a second standalone cloning skill would split guidance and make verification behavior drift.
-surface-class: "historical"
-domain: "creator-runtime"
-audience: "internal"

package/skills/wiki/SKILL.md

@@ -59,6 +59,3 @@ At session end, discoveries can be captured as `session-log-*` pages. Configure
 ## Hard Constraints
 - No vector embeddings — query uses keyword + tag matching only
 - Wiki files remain local project state under `.rcs/wiki/`
-surface-class: "operator"
-domain: "creator-runtime"
-audience: "operator"

package/skills/worker/SKILL.md

@@ -9,7 +9,7 @@ artifact-type: "skill"
 
 # Worker Skill
 
-> Legacy compatibility surface. Public creator language should prefer `crew` / coordinated execution terminology. This file remains the internal worker-runtime protocol surface for tmux-managed crew panes.
+> Internal worker-runtime surface. Active creator language should prefer `crew` / coordinated execution terminology. This file exists so tmux-managed crew panes follow one strict protocol.
 
 This skill is for a Codex session that was started as a coordinated crew worker pane under `rcs team` / `$team`.
 
@@ -110,6 +110,3 @@ Crew worker sessions should treat team state + CLI interop as the source of trut
 ## Shutdown
 
 If the crew lead sends a shutdown request, follow the shutdown inbox instructions exactly, write your shutdown ack file, then exit the Codex session.
-surface-class: "internal"
-domain: "creator-runtime"
-audience: "internal"

package/src/scripts/publish-github-package.ts

@@ -0,0 +1,94 @@
+#!/usr/bin/env node
+import { readFileSync, writeFileSync } from 'node:fs';
+import { basename, join } from 'node:path';
+import { spawnSync } from 'node:child_process';
+
+interface PackageJson {
+  name?: string;
+  version?: string;
+  publishConfig?: Record<string, unknown>;
+  repository?: string | { url?: string; [key: string]: unknown };
+}
+
+interface PackageLockJson {
+  name?: string;
+  version?: string;
+  packages?: Record<string, { name?: string; version?: string; [key: string]: unknown }>;
+}
+
+function normalizeScopeOwner(owner: string): string {
+  return owner.trim().toLowerCase();
+}
+
+function resolveRepositoryUrl(pkg: PackageJson): string {
+  if (typeof pkg.repository === 'string') return pkg.repository;
+  if (pkg.repository && typeof pkg.repository === 'object' && typeof pkg.repository.url === 'string') {
+    return pkg.repository.url;
+  }
+  throw new Error('github_packages_publish_requires_repository_url');
+}
+
+function resolveGitHubOwner(pkg: PackageJson): string {
+  const repositoryUrl = resolveRepositoryUrl(pkg);
+  const match = repositoryUrl.match(/github\.com[:/](.+?)\/(.+?)(?:\.git)?$/i);
+  if (!match?.[1]) {
+    throw new Error(`unable_to_resolve_github_owner_from_repository:${repositoryUrl}`);
+  }
+  return normalizeScopeOwner(match[1]);
+}
+
+function resolvePackageBaseName(pkg: PackageJson): string {
+  const name = (pkg.name ?? '').trim();
+  if (!name) throw new Error('github_packages_publish_requires_package_name');
+  const last = name.split('/').at(-1);
+  if (!last) throw new Error('github_packages_publish_requires_package_name');
+  return last;
+}
+
+function buildGitHubPackagesName(pkg: PackageJson): string {
+  return `@${resolveGitHubOwner(pkg)}/${resolvePackageBaseName(pkg)}`;
+}
+
+function main(): void {
+  const root = process.cwd();
+  const dryRun = process.argv.includes('--dry-run');
+  const packagePath = join(root, 'package.json');
+  const lockPath = join(root, 'package-lock.json');
+
+  const originalPackage = readFileSync(packagePath, 'utf8');
+  const originalLock = readFileSync(lockPath, 'utf8');
+
+  const pkg = JSON.parse(originalPackage) as PackageJson;
+  const lock = JSON.parse(originalLock) as PackageLockJson;
+  const githubPackageName = buildGitHubPackagesName(pkg);
+
+  pkg.name = githubPackageName;
+  pkg.publishConfig = {
+    ...(pkg.publishConfig ?? {}),
+    registry: 'https://npm.pkg.github.com',
+  };
+
+  lock.name = githubPackageName;
+  if (lock.packages?.['']) {
+    lock.packages[''].name = githubPackageName;
+  }
+
+  writeFileSync(packagePath, `${JSON.stringify(pkg, null, 2)}\n`);
+  writeFileSync(lockPath, `${JSON.stringify(lock, null, 2)}\n`);
+
+  try {
+    const args = ['publish', ...(dryRun ? ['--dry-run'] : [])];
+    const result = spawnSync('npm', args, {
+      cwd: root,
+      stdio: 'inherit',
+      env: process.env,
+    });
+    if (result.error) throw result.error;
+    process.exit(result.status ?? 1);
+  } finally {
+    writeFileSync(packagePath, originalPackage);
+    writeFileSync(lockPath, originalLock);
+  }
+}
+
+main();

package/templates/roblox/llm-provider-routing.json

@@ -0,0 +1,42 @@
+{
+  "providers": {
+    "openrouter": {
+      "base_url": "https://openrouter.ai/api/v1",
+      "api_format": "openai-compatible",
+      "env_key": "OPENROUTER_API_KEY",
+      "capabilities": [
+        "chat",
+        "reasoning"
+      ],
+      "label": "OpenRouter"
+    },
+    "anthropic_proxy": {
+      "base_url": "https://proxy.example.com/anthropic",
+      "api_format": "anthropic",
+      "env_key": "ANTHROPIC_PROXY_KEY",
+      "capabilities": [
+        "chat",
+        "long-context"
+      ],
+      "label": "Anthropic Proxy"
+    }
+  },
+  "routing": {
+    "default_provider": "openrouter",
+    "mode_providers": {
+      "forge": "openrouter",
+      "team": "anthropic_proxy"
+    },
+    "fallback_providers": [
+      "openrouter"
+    ],
+    "mode_fallback_providers": {
+      "forge": [
+        "openrouter",
+        "anthropic_proxy"
+      ]
+    },
+    "hot_swap": true,
+    "failover": true
+  }
+}

package/templates/roblox/reference-sources.md

@@ -0,0 +1,33 @@
+# Roblox Reference Sources
+
+Use this inventory when a Roblox task needs external grounding before implementation.
+
+## Official platform truth
+
+- `https://github.com/Roblox/creator-docs`
+- `https://gitmcp.io/Roblox/creator-docs`
+
+## Roblox implementation guidance
+
+- `https://github.com/sentinelcore/roblox-skills`
+- `https://gitmcp.io/sentinelcore/roblox-skills`
+- `https://github.com/greedychipmunk/agent-skills/tree/main/roblox-game-developer`
+- `https://github.com/omer-metin/skills-for-antigravity/tree/main/skills/roblox-development`
+- `https://github.com/dig1t/skills`
+- `https://github.com/Corecii/Devprod`
+- `https://gitmcp.io/Corecii/Devprod`
+
+## Raw script / corpus support
+
+- `https://github.com/retpirato/Roblox-Scripts`
+- `https://gitmcp.io/retpirato/Roblox-Scripts`
+- `https://gitmcp.io/frosteen/Roblox_LUA_Weapon_Scripts`
+- `https://gitmcp.io/uhub/awesome-lua`
+- `https://gitmcp.io/LewisJEllis/awesome-lua`
+- `https://gitmcp.io/forhappy/awesome-lua`
+- `http://lua-users.org/wiki/SampleCode`
+
+Rules:
+- official docs define correctness
+- skill repos improve implementation awareness
+- corpora are not trusted production code