mp-ai-module-controller/manifest.llm.json

{
  "schema_version": "1.0",
  "name": "mp-ai-module-controller",
  "description": "Rust CLI controller for llama.cpp-backed local action dispatch.",
  "project_type": "cli",
  "language": "rust",
  "package_manager": "cargo",
  "entrypoints": {
    "main": "src/main.rs",
    "action_registry": "src/registry.rs",
    "dictionary_generator": "src/dictionary.rs",
    "llm_client": "src/llm.rs",
    "dispatcher": "src/dispatcher.rs",
    "logger_script": "src/scripts/logger.rs",
    "codex_environment": ".codex/environments/environment.toml"
  },
  "commands": {
    "build": "cargo build",
    "test": "cargo test",
    "generate_dictionary": "cargo run -- dictionary generate",
    "serve": "cargo run -- serve",
    "query": "cargo run -- query \"log hello from codex\"",
    "run_logger": "cargo run -- run logger --payload '{\"message\":\"hello from codex\"}'",
    "run_calculate": "cargo run -- run calculate --payload '{\"expression\":\"2 + 2 * 3\"}'",
    "run_to_slug": "cargo run -- run to_slug --payload '{\"text\":\"Hello, World!\"}'",
    "run_write_note": "cargo run -- run write_note --payload '{\"filename\":\"hello.txt\",\"content\":\"hello\"}'",
    "codex_setup": "scripts/codex-setup.sh",
    "codex_logger_smoke": "scripts/codex-run-logger.sh",
    "precommit_check": "scripts/precommit-check.sh",
    "training_data_dry_run": "python3 scripts/generate_training_data.py --dry-run",
    "generate_training_data": "python3 scripts/generate_training_data.py",
    "deploy_training_data": "scripts/deploy_training_data.sh",
    "deploy_training_data_with_mimir": "ENABLE_MIMIR_SDG=1 scripts/deploy_training_data.sh",
    "install_hooks": "git config core.hooksPath .githooks"
  },
  "environment": {
    "example_file": ".env.example",
    "variables": [
      {
        "name": "LLAMA_MODEL_PATH",
        "required": true,
        "secret": false,
        "description": "Absolute path to a local GGUF model used by llama-server."
      },
      {
        "name": "LLAMA_HOST",
        "required": false,
        "secret": false,
        "default": "127.0.0.1",
        "description": "Host for the local llama.cpp server."
      },
      {
        "name": "LLAMA_PORT",
        "required": false,
        "secret": false,
        "default": "8080",
        "description": "Port for the local llama.cpp server."
      },
      {
        "name": "LLAMA_CONTEXT_SIZE",
        "required": false,
        "secret": false,
        "description": "Optional context size passed to llama-server as --ctx-size."
      },
      {
        "name": "LLAMA_EXTRA_ARGS",
        "required": false,
        "secret": false,
        "description": "Optional shell-split extra arguments appended to llama-server."
      },
      {
        "name": "CONTROLLER_LOG_LEVEL",
        "required": false,
        "secret": false,
        "default": "info",
        "description": "Tracing log level for the controller."
      },
      {
        "name": "ENABLE_MIMIR_SDG",
        "required": false,
        "secret": false,
        "default": "0",
        "description": "Set to 1 to run Mimir/Data Designer expansion during training-data deploys."
      },
      {
        "name": "MIMIR_SDG_HOST",
        "required": false,
        "secret": false,
        "default": "100.80.52.47",
        "description": "Mimir host used for Data Designer expansion."
      },
      {
        "name": "MIMIR_SDG_USER",
        "required": false,
        "secret": false,
        "default": "aaron-pressey",
        "description": "SSH user for Mimir Data Designer expansion."
      },
      {
        "name": "MIMIR_SDG_KEY",
        "required": false,
        "secret": false,
        "default": "~/.ssh/silma_orson_ed25519",
        "description": "SSH key for Mimir Data Designer expansion."
      },
      {
        "name": "MIMIR_SDG_MODEL_ALIAS",
        "required": false,
        "secret": false,
        "default": "nvidia-text",
        "description": "Data Designer model alias used for SDG expansion."
      },
      {
        "name": "MIMIR_SDG_NUM_RECORDS",
        "required": false,
        "secret": false,
        "default": "1000",
        "description": "Number of enriched SDG records requested from Data Designer."
      },
      {
        "name": "NVIDIA_API_KEY",
        "required": false,
        "secret": true,
        "description": "Provider credential for local Data Designer usage. Keep it in ignored .env files or Mimir's Data Designer env file."
      }
    ]
  },
  "llama_cpp": {
    "server_binary": "/opt/homebrew/bin/llama-server",
    "fallback_binary": "llama-server",
    "api": "OpenAI-compatible /v1/chat/completions",
    "response_contract": {
      "format": "verbose JSON or compact codebook JSON",
      "example": {
        "action_id": "logger",
        "payload": {
          "message": "hello"
        }
      },
      "compact_example": {
        "v": "1.0.0",
        "c": "<codebook_checksum>",
        "a": "A001",
        "p": {
          "F001": "hello"
        }
      }
    }
  },
  "dictionary": {
    "source_of_truth": "src/registry.rs",
    "static_base": "dictionary/static-base.json",
    "generated_jsonl": "dictionary/actions.jsonl",
    "generated_index": "dictionary/actions.index.json",
    "generated_codebook": "dictionary/model.codebook.json",
    "generated_examples": "dictionary/model.examples.jsonl",
    "dictionary_version": "1.0.0",
    "metadata": [
      "registry_checksum",
      "static_base_checksum",
      "codebook_checksum"
    ],
    "runtime_policy": "Verbose outputs require exact action_id matches; compact outputs require matching version/checksum and exact codebook expansion before dispatch."
  },
  "actions": [
    {
      "action_id": "calculate",
      "suite": "Math & Conversion",
      "payload_example": {
        "expression": "2 + 2 * 3"
      }
    },
    {
      "action_id": "convert_units",
      "suite": "Math & Conversion",
      "payload_example": {
        "value": 100,
        "from": "km",
        "to": "miles"
      }
    },
    {
      "action_id": "generate_uuid",
      "suite": "Math & Conversion",
      "payload_example": {}
    },
    {
      "action_id": "random_number",
      "suite": "Math & Conversion",
      "payload_example": {
        "min": 1,
        "max": 100
      }
    },
    {
      "action_id": "count_words",
      "suite": "Text Utilities",
      "payload_example": {
        "text": "hello from codex"
      }
    },
    {
      "action_id": "to_slug",
      "suite": "Text Utilities",
      "payload_example": {
        "text": "Hello, World!"
      }
    },
    {
      "action_id": "hash_string",
      "suite": "Text Utilities",
      "payload_example": {
        "text": "hello",
        "algorithm": "sha256"
      }
    },
    {
      "action_id": "truncate_text",
      "suite": "Text Utilities",
      "payload_example": {
        "text": "hello from codex",
        "max_chars": 5
      }
    },
    {
      "action_id": "write_note",
      "suite": "File Operations",
      "payload_example": {
        "filename": "hello.txt",
        "content": "hello"
      },
      "sandbox": "notes/"
    },
    {
      "action_id": "append_to_file",
      "suite": "File Operations",
      "payload_example": {
        "filename": "hello.txt",
        "content": " world"
      },
      "sandbox": "notes/"
    },
    {
      "action_id": "read_file",
      "suite": "File Operations",
      "payload_example": {
        "filename": "hello.txt"
      },
      "sandbox": "notes/"
    },
    {
      "action_id": "delete_file",
      "suite": "File Operations",
      "payload_example": {
        "filename": "hello.txt"
      },
      "sandbox": "notes/"
    },
    {
      "action_id": "logger",
      "suite": "Logger",
      "script": "src/scripts/logger.rs",
      "description": "Proof-of-concept script that writes structured JSONL log records.",
      "payload_example": {
        "message": "hello from codex"
      },
      "output": "logs/controller-actions.jsonl"
    }
  ],
  "generated_paths": [
    "target/",
    "logs/",
    "runtime/",
    "notes/",
    "dictionary/actions.jsonl",
    "dictionary/actions.index.json",
    "dictionary/model.codebook.json",
    "dictionary/model.examples.jsonl"
  ],
  "sdg_bridge": {
    "design_doc": "SDG_BRIDGE_PLAN.md",
    "script": "scripts/generate_training_data.py",
    "deploy_script": "scripts/deploy_training_data.sh",
    "custom_local_llm_root": "/Volumes/Zoe/custom-local-llm",
    "default_output_dir": "/Volumes/Zoe/custom-local-llm/training-data/",
    "default_manifest": "/Volumes/Zoe/custom-local-llm/manifest.llm.json",
    "dataset_prefix": "module_controller_intents",
    "schema_profile": "instruction",
    "dry_run": "python3 scripts/generate_training_data.py --dry-run",
    "optional_sdg_host": "mimir (100.80.52.47)",
    "sdg_engine": "data-designer",
    "default_sdg_model_alias": "nvidia-text",
    "sdg_model_alias_override_env": "MIMIR_SDG_MODEL_ALIAS",
    "sdg_invocation": "ssh to Mimir and run Data Designer CLI",
    "sdg_env_file": "/home/aaron-pressey/.config/home-grown-llm-data/data-designer.env",
    "expanded_raw_file": "expanded.raw.jsonl",
    "expanded_normalized_file": "expanded.jsonl",
    "woodpecker_deploy_command": "ENABLE_MIMIR_SDG=1 scripts/deploy_training_data.sh",
    "seed_data_note": "train.jsonl and validation.jsonl are deterministic seeds; use expanded.jsonl for enriched model training when available."
  },
  "ci": {
    "codex_environment": ".codex/environments/environment.toml",
    "gitea_actions": ".gitea/workflows/ci.yml",
    "woodpecker": ".woodpecker.yml",
    "woodpecker_deploy": "scripts/deploy_training_data.sh",
    "precommit_hook": ".githooks/pre-commit",
    "shared_check": "scripts/precommit-check.sh",
    "checks": [
      "cargo fmt --check",
      "cargo test",
      "cargo run -- dictionary generate",
      "test -s dictionary/actions.jsonl",
      "test -s dictionary/actions.index.json",
      "test -s dictionary/model.codebook.json",
      "test -s dictionary/model.examples.jsonl",
      "test -s dictionary/static-base.json",
      "python3 -m unittest tests/test_generate_training_data.py",
      "python3 scripts/generate_training_data.py --dry-run"
    ]
  },
  "agent_guidance": "See AGENTS.md. Keep src/registry.rs as the source of truth, regenerate dictionary files after action changes, and reject non-exact LLM action outputs."
}