tailtest_scenario_plan
Generates structured scaffolding that an agent uses to write a SCENARIO PLAN, detailing language, framework, depth, adversarial count, baseline scenarios, and test file path for automated test creation.
Instructions
Return structured scaffolding the agent uses to write its SCENARIO PLAN: language, framework, depth, R15 adversarial count requirement, language and framework baseline scenarios, test file path, and prose instructions. The agent uses this scaffolding to compose the actual SCENARIO PLAN scenario lines.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| file_path | Yes | Relative or absolute path to the source file under test. | |
| project_root | No | Project root directory. Defaults to the current working directory. |
Implementation Reference
- The main handler function `scenario_plan()` that executes the tool logic. It reads depth from config, detects language/framework, computes scenario counts and adversarial requirements, and returns structured scaffolding for the agent's SCENARIO PLAN.
def scenario_plan(file_path: str, project_root: str | None = None) -> dict[str, Any]: """Return structured scaffolding the agent uses to write its SCENARIO PLAN. Args: file_path: relative or absolute path to the source file. project_root: project root for reading config.json. Defaults to cwd. Returns: Dict with: file_path, language, framework, depth, scenario_count_target, adversarial_count_required, adversarial_categories, language_baseline, framework_baseline, test_file_path, instructions. """ project_root = project_root or os.getcwd() language = detect_language(file_path) or "unknown" framework = _detect_framework(language, file_path, project_root) depth = _read_depth(project_root) count_min, count_max = SCENARIO_COUNT_BY_DEPTH[depth] adv_required = ADVERSARIAL_BY_DEPTH[depth] test_path = _test_file_path(file_path, language, framework) instructions = ( f"Generate a SCENARIO PLAN for {file_path}. " f"Depth is {depth}: produce {count_min} to {count_max} scenarios total. " f"R15 requires at least {adv_required} adversarial scenarios labeled " f"[adversarial: <category>]. Pick categories from the 8-category list " f"that genuinely apply to this file; document any skipped category with " f"a reason. Include the language baseline scenarios. " ) if framework: instructions += ( f"Include the {framework} framework baseline scenarios on top of the " f"language baseline. " ) if depth == "simple": instructions += ( "Note: at depth: simple, R15 does not apply -- generate happy-path " "scenarios only. " ) return { "file_path": file_path, "language": language, "framework": framework, "depth": depth, "scenario_count_target": [count_min, count_max], "adversarial_count_required": adv_required, "adversarial_categories": ADVERSARIAL_CATEGORIES, "language_baseline": LANGUAGE_BASELINE.get(language, []), "framework_baseline": _framework_baseline(framework), "test_file_path": test_path, "instructions": instructions, } - mcp_server/src/tailtest_mcp/server.py:37-60 (registration)Registration of 'tailtest_scenario_plan' as a Tool in the MCP server's list_tools() with input schema (file_path required, project_root optional).
Tool( name="tailtest_scenario_plan", description=( "Return structured scaffolding the agent uses to write its SCENARIO PLAN: " "language, framework, depth, R15 adversarial count requirement, language and " "framework baseline scenarios, test file path, and prose instructions. The agent " "uses this scaffolding to compose the actual SCENARIO PLAN scenario lines." ), inputSchema={ "type": "object", "properties": { "file_path": { "type": "string", "description": "Relative or absolute path to the source file under test.", }, "project_root": { "type": "string", "description": "Project root directory. Defaults to the current working directory.", }, }, "required": ["file_path"], "additionalProperties": False, }, ), - mcp_server/src/tailtest_mcp/server.py:157-165 (registration)Dispatch/call handler in server.py that imports and invokes scenario_plan() when the tool name is 'tailtest_scenario_plan'.
if name == "tailtest_scenario_plan": from .tools.scenario_plan import scenario_plan import json as _json result = scenario_plan( file_path=arguments["file_path"], project_root=arguments.get("project_root"), ) return [TextContent(type="text", text=_json.dumps(result, indent=2))] - Helper _read_depth() reads the .tailtest/config.json depth setting.
def _read_depth(project_root: str) -> str: """Read depth from .tailtest/config.json. Defaults to 'standard'.""" config_path = os.path.join(project_root, ".tailtest", "config.json") if os.path.exists(config_path): try: with open(config_path) as f: cfg = json.load(f) value = cfg.get("depth") if value in ADVERSARIAL_BY_DEPTH: return value except (json.JSONDecodeError, OSError): pass return "standard" - Helper _framework_baseline() returns framework-specific baseline scenarios (flask, fastapi, django).
def _framework_baseline(framework: str | None) -> list[str]: """Framework baseline scenarios from R2 templates.""" if framework is None: return [] if framework == "flask": return [ "Route returns 200 on valid path", "404 on unknown route", "Blueprint registration binds the correct prefix", "test_client fixture used within app context", "Validation rejects bad input", ] if framework == "fastapi": return [ "Valid request body returns expected response", "Missing required field returns 422", "Wrong field type returns 422", "Dependency override works in test (app.dependency_overrides)", ] if framework == "django": return [ "Request with valid auth", "Request without auth (expect 403/redirect)", "Model field validation rejects invalid data", "URL routes to the correct view", ] return []