tribunal-kit 9.2.4 → 9.2.6

package/.agent/mcp_config.json

@@ -1,32 +1,32 @@
-{
-  "mcpServers": {
-    "context7": {
-      "command": "npx",
-      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_CONTEXT7_API_KEY"]
-    },
-    "shadcn": {
-      "command": "npx",
-      "args": ["shadcn@latest", "mcp"]
-    }
-  }
-}
-// ━━━ Setup Instructions ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
-//
-// Copy this file to your Antigravity MCP config location:
-//   Windows: %USERPROFILE%\.gemini\antigravity\mcp_config.json
-//   macOS:   ~/.gemini/antigravity/mcp_config.json
-//
-// Replace YOUR_CONTEXT7_API_KEY with your real key from:
-//   https://upstash.com
-//
-// To add more MCP servers, append entries to "mcpServers":
-// {
-//   "my-server": {
-//     "command": "npx",
-//     "args": ["-y", "my-mcp-package"]
-//   }
-// }
-//
-// What these servers provide:
-//   context7  → Live documentation lookup (framework docs, APIs)
-//   shadcn    → shadcn/ui component integration
+{
+  "mcpServers": {
+    "context7": {
+      "command": "npx",
+      "args": ["-y", "@upstash/context7-mcp", "--api-key", "YOUR_CONTEXT7_API_KEY"]
+    },
+    "shadcn": {
+      "command": "npx",
+      "args": ["shadcn@latest", "mcp"]
+    }
+  }
+}
+// ━━━ Setup Instructions ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+//
+// Copy this file to your Antigravity MCP config location:
+//   Windows: %USERPROFILE%\.gemini\antigravity\mcp_config.json
+//   macOS:   ~/.gemini/antigravity/mcp_config.json
+//
+// Replace YOUR_CONTEXT7_API_KEY with your real key from:
+//   https://upstash.com
+//
+// To add more MCP servers, append entries to "mcpServers":
+// {
+//   "my-server": {
+//     "command": "npx",
+//     "args": ["-y", "my-mcp-package"]
+//   }
+// }
+//
+// What these servers provide:
+//   context7  → Live documentation lookup (framework docs, APIs)
+//   shadcn    → shadcn/ui component integration

package/.agent/patterns/generator.md

@@ -1,11 +1,11 @@
-# Generator Pattern
-
-**Purpose**: Produce structured output by filling a reusable template governed by quality rules.
-
-## Protocol
-
-When a skill inherits this pattern, the agent is tasked with producing a specific formatted artifact (like a configuration file, documentation page, or scaffolding code).
-
-1. **Template Retrieval**: Locate and strictly adhere to the provided template structure (the "assets") defined by the specific skill.
-2. **Constraint Application**: Apply all quality rules and constraints (the "references") required by the skill while fleshing out the template.
-3. **No Halucination Formatting**: Do not invent new sections, alter the required Markdown/JSON structure, or add unauthorized commentary unless it fits directly into the predefined template slots.
+# Generator Pattern
+
+**Purpose**: Produce structured output by filling a reusable template governed by quality rules.
+
+## Protocol
+
+When a skill inherits this pattern, the agent is tasked with producing a specific formatted artifact (like a configuration file, documentation page, or scaffolding code).
+
+1. **Template Retrieval**: Locate and strictly adhere to the provided template structure (the "assets") defined by the specific skill.
+2. **Constraint Application**: Apply all quality rules and constraints (the "references") required by the skill while fleshing out the template.
+3. **No Halucination Formatting**: Do not invent new sections, alter the required Markdown/JSON structure, or add unauthorized commentary unless it fits directly into the predefined template slots.

package/.agent/patterns/inversion.md

@@ -1,14 +1,14 @@
-# Inversion Pattern
-
-**Purpose**: Interview the user before taking action.
-
-## Protocol
-
-When a skill inherits this pattern, you MUST NOT proceed with execution immediately. Instead, rely on the "Socratic Gate". You must pause and ask the user questions using the following structured phases:
-
-1. **Identify Missing Context**: Evaluate the user's prompt against what is absolutely necessary to execute the skill.
-2. **Phase 1 (Goal & Constraints)**: Ask the user about the real outcome and any hard constraints.
-3. **Phase 2 (Out of Scope)**: Confirm what should explicitly NOT be done.
-4. **Phase 3 (Done Condition)**: Verify how you will know the task is completed.
-
-You must receive explicit answers or a "do your best" override before writing code or executing substantive actions.
+# Inversion Pattern
+
+**Purpose**: Interview the user before taking action.
+
+## Protocol
+
+When a skill inherits this pattern, you MUST NOT proceed with execution immediately. Instead, rely on the "Socratic Gate". You must pause and ask the user questions using the following structured phases:
+
+1. **Identify Missing Context**: Evaluate the user's prompt against what is absolutely necessary to execute the skill.
+2. **Phase 1 (Goal & Constraints)**: Ask the user about the real outcome and any hard constraints.
+3. **Phase 2 (Out of Scope)**: Confirm what should explicitly NOT be done.
+4. **Phase 3 (Done Condition)**: Verify how you will know the task is completed.
+
+You must receive explicit answers or a "do your best" override before writing code or executing substantive actions.

package/.agent/patterns/pipeline.md

@@ -1,11 +1,11 @@
-# Pipeline Pattern
-
-**Purpose**: Link multiple execution steps together with explicit validation gates between them.
-
-## Protocol
-
-When a skill inherits this pattern, the agent must execute its instructions sequentially and rigidly.
-
-1. **Step-by-Step Execution**: You must not skip steps or combine multiple distinct phases into a single massive generative output.
-2. **Validation Gates**: After completing Step N, you must validate that the output of Step N meets its success criteria before moving to Step N+1.
-3. **Halting**: If any gate fails validation, you must HALT the pipeline and either initiate an Error Recovery Protocol or report the failure to the user. Do not proceed with subsequent steps with broken inputs.
+# Pipeline Pattern
+
+**Purpose**: Link multiple execution steps together with explicit validation gates between them.
+
+## Protocol
+
+When a skill inherits this pattern, the agent must execute its instructions sequentially and rigidly.
+
+1. **Step-by-Step Execution**: You must not skip steps or combine multiple distinct phases into a single massive generative output.
+2. **Validation Gates**: After completing Step N, you must validate that the output of Step N meets its success criteria before moving to Step N+1.
+3. **Halting**: If any gate fails validation, you must HALT the pipeline and either initiate an Error Recovery Protocol or report the failure to the user. Do not proceed with subsequent steps with broken inputs.

package/.agent/patterns/reviewer.md

@@ -1,15 +1,15 @@
-# Reviewer Pattern
-
-**Purpose**: Evaluate code or content against a strict external checklist.
-
-## Protocol
-
-When a skill inherits this pattern, the agent assumes the role of an evaluator. Do NOT generate novel content or fix the problem automatically unless explicitly instructed.
-
-1. **Checklist Enforcement**: You must read the evaluation checklist provided in the specific skill.
-2. **Review Output**: For every item in the checklist, determine if it passes or fails.
-3. **Severity Grading**: Group all findings by severity:
-   - **Critical**: Must fix before proceeding (e.g. security violations, build errors)
-   - **Warning**: Should fix (e.g. best practice violations, performance risks)
-   - **Info**: Stylistic or minor suggestions
-4. **Separation of Concerns**: Only evaluate the "what" (the checklist) based on the "how" (this standard format). Do not blur your own opinions into the checklist constraints.
+# Reviewer Pattern
+
+**Purpose**: Evaluate code or content against a strict external checklist.
+
+## Protocol
+
+When a skill inherits this pattern, the agent assumes the role of an evaluator. Do NOT generate novel content or fix the problem automatically unless explicitly instructed.
+
+1. **Checklist Enforcement**: You must read the evaluation checklist provided in the specific skill.
+2. **Review Output**: For every item in the checklist, determine if it passes or fails.
+3. **Severity Grading**: Group all findings by severity:
+   - **Critical**: Must fix before proceeding (e.g. security violations, build errors)
+   - **Warning**: Should fix (e.g. best practice violations, performance risks)
+   - **Info**: Stylistic or minor suggestions
+4. **Separation of Concerns**: Only evaluate the "what" (the checklist) based on the "how" (this standard format). Do not blur your own opinions into the checklist constraints.

package/.agent/patterns/tool-wrapper.md

@@ -1,11 +1,11 @@
-# Tool Wrapper Pattern
-
-**Purpose**: Package an external library's or CLI tool's conventions as on-demand, executable knowledge.
-
-## Protocol
-
-When a skill inherits this pattern, the agent MUST NOT guess how to use the target tool. You are acting strictly as a wrapper for this specific utility.
-
-1. **Consult References**: Read the provided documentation, usage examples, or reference notes in the skill definitions BEFORE issuing any commands.
-2. **Strict Adherence**: Follow the rules defined in the skill exactly as written. Do not improvise flags, parameters, or endpoints that are not explicitly authorized by the reference.
-3. **Command Execution**: If the tool is a CLI command or Python script (e.g. `test_runner.py`), construct the command accurately based solely on the referenced conventions, execute it, and report the direct output.
+# Tool Wrapper Pattern
+
+**Purpose**: Package an external library's or CLI tool's conventions as on-demand, executable knowledge.
+
+## Protocol
+
+When a skill inherits this pattern, the agent MUST NOT guess how to use the target tool. You are acting strictly as a wrapper for this specific utility.
+
+1. **Consult References**: Read the provided documentation, usage examples, or reference notes in the skill definitions BEFORE issuing any commands.
+2. **Strict Adherence**: Follow the rules defined in the skill exactly as written. Do not improvise flags, parameters, or endpoints that are not explicitly authorized by the reference.
+3. **Command Execution**: If the tool is a CLI command or Python script (e.g. `test_runner.py`), construct the command accurately based solely on the referenced conventions, execute it, and report the direct output.

package/.agent/plans/implementation_plan.md

@@ -0,0 +1,55 @@
+# Tribunal Kit Master Implementation Plan (ADR-Lite)
+
+## 1. Objective Context
+
+Modernize the existing Tribunal Kit framework by integrating Durable Session Logs, a secure Agent Syscall Abstraction (Tribunal OS), Tribunal Instincts (persistent evidence-backed memory), Dynamic Team Mode (DAG-based subagent delegation), and Built-in MCP Integration. The goal is a more reliable, context-efficient, and secure governance harness.
+
+## 2. Architectural Handoff
+
+- **Stack:** Node.js (CommonJS/ESM hybrid), standard filesystem (`fs`, `path`), Child Processes (`child_process`).
+- **Storage:** Local JSON/JSONL files (no external databases required).
+- **Constraints:** Backwards compatible with existing `.agent` directory structures, skills, and slash commands. Strict security boundary for tool execution (no raw shell executions by agents).
+
+## 3. Task-Level Interface Contracts
+
+- **`session_logger.js`**:
+  - _Produces_: `logEvent(type, payload)` -> `string` (Event ID). `readEvents()` -> `Array<object>`.
+- **`syscall_registry.js`**:
+  - _Produces_: `executeSyscall(toolName, args, context)` -> `Promise<{stdout, stderr, code}>`.
+- **`memory_archivist.js`**:
+  - _Consumes_: `session_logger.readEvents()`.
+  - _Produces_: `instincts.json` schemas.
+- **`swarm_dispatcher.js`**:
+  - _Consumes_: `harness_manager` tool permissions.
+  - _Produces_: DAG execution result (`allSettled` parallel evaluation).
+- **`mcp_server.js`**:
+  - _Consumes_: `stdio` (JSON-RPC 2.0).
+  - _Produces_: Standard MCP responses routing to `syscall_registry`.
+
+## 4. The Zero-Placeholder Invariant
+
+All implementations will contain concrete Node.js logic with explicit filesystem checks, child process error boundaries, and standard Tribunal Kit `require('./_colors')` integrations. No "TBD" comments are allowed.
+
+## 5. Dependency Tree Execution Order
+
+1. **Foundation (Wave 1):** Session Logger and Syscall Registry. Cannot execute safe workflows without an event log and isolated execution environment.
+2. **Memory (Wave 2):** Tribunal Instincts. Depends on the Session Logger to extract insights.
+3. **Orchestration (Wave 3):** Team Mode & DAGs. Depends on Syscalls for execution isolation.
+4. **Integration (Wave 4):** MCP Server. Depends on the existing registry and orchestration engine to expose tools externally.
+
+## 6. File Blueprint
+
+- `[NEW]` `.agent/scripts/session_logger.js`
+- `[NEW]` `.agent/scripts/memory_archivist.js`
+- `[NEW]` `.agent/scripts/syscall_registry.js`
+- `[NEW]` `.agent/scripts/harness_manager.js`
+- `[NEW]` `bin/mcp-server.js`
+- `[MODIFY]` `.agent/scripts/pipeline_engine.js`
+- `[MODIFY]` `.agent/scripts/swarm_dispatcher.js`
+- `[MODIFY]` `test/unit/audit_release.test.js`
+
+## 7. Verification Protocol
+
+- **Linter & Security:** Code must pass `node .agent/scripts/checklist.js .` with 0 ESLint warnings (`no-unused-vars` strictly handled via `_` prefix) and 0 security scan alerts.
+- **Test Suite:** `npm run test` must exit `0`, proving DAG topological sorts, MCP request handling, and backward compatibility with `pipeline_engine.js`.
+- **Integration:** Booting the `mcp-server.js` manually must respond to `{"jsonrpc":"2.0","method":"resources/list","id":1}`.

package/.agent/plans/task.md

@@ -0,0 +1,41 @@
+# Current Objective: Agent Harness Modernization & Orchestration
+
+## Pre-Flight
+
+- [x] Dump existing `.agent` directory state
+- [x] Verify current tests pass (Baseline health via `/audit`)
+
+## Wave 1 (Durable Session Log & Syscall execution)
+
+- [ ] Implement `session_logger.js` (Event-driven persistent JSONL storage)
+- [ ] Implement `syscall_registry.js` (Centralized tool permissions and execution boundaries)
+- [ ] Implement `harness_manager.js` (Syscall integration and process separation)
+- [ ] Update `pipeline_engine.js` to log events via `session_logger`
+- [ ] Run `npm test` and `/audit`
+
+## Wave 2 (Tribunal Instincts & Continuous Memory)
+
+- [ ] Implement `memory_archivist.js` (Background extraction of lessons into `instincts.json`)
+- [ ] Define instinct lifecycle and validation logic
+- [ ] Inject instincts into `context_broker.js`
+- [ ] Run `npm test` and `/audit`
+
+## Wave 3 (Dynamic Team Mode & DAG Orchestration)
+
+- [ ] Refactor `swarm_dispatcher.js` to support topological sorting (`executeDAG`)
+- [ ] Introduce DAG-based workflow parsing (e.g. `kernel-scheduler.md` concepts)
+- [ ] Implement failure recovery and dependency skipping
+- [ ] Run `npm test` and `/audit`
+
+## Wave 4 (MCP Integration)
+
+- [ ] Implement `mcp_server.js` (JSON-RPC stdio adapter for Tribunal syscalls)
+- [ ] Expose resource listing (`resources/list`) and prompt listing (`prompts/list`)
+- [ ] Add tool execution mapping to `syscall_registry.js`
+- [ ] Run `npm test` and `/audit`
+
+## Wave 5 (Finalization & Security Audit)
+
+- [ ] Complete final system audit (resolving linting, test counts, unused vars)
+- [ ] Generate final integration tests for MCP and Syscall registries
+- [ ] Produce `final-tribunal-kit-enhancement-audit.md`