requirements-agent 0.0.1__tar.gz

requirements_agent-0.0.1/PKG-INFO

@@ -0,0 +1,166 @@
+Metadata-Version: 2.4
+Name: requirements-agent
+Version: 0.0.1
+Summary: Self-contained Spec Circuit Requirements Agent (CLI + Textual TUI)
+Author: CLI-Agents
+License: MIT
+Requires-Python: >=3.9
+Description-Content-Type: text/markdown
+Requires-Dist: typer>=0.12
+Requires-Dist: rich>=13.7
+Requires-Dist: textual>=0.80
+Requires-Dist: openai>=1.40
+Requires-Dist: httpx>=0.27
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: tomli>=2.0; python_version < "3.11"
+Requires-Dist: fpdf2>=2.7
+Requires-Dist: markdown>=3.5
+Provides-Extra: pdf
+Requires-Dist: weasyprint>=62.0; extra == "pdf"
+Provides-Extra: dev
+Requires-Dist: pytest>=8.0; extra == "dev"
+Requires-Dist: pytest-asyncio>=0.23; extra == "dev"
+
+# Requirements Agent
+
+Self-contained **Spec Circuit** requirements agent: multi-persona BMAD review → quality score → refine → BRD → Markdown/PDF.
+
+No Atlas / FastAPI / Mongo dependency. Configure LLMs in JSON, agent settings in YAML, run from CLI or Textual TUI.
+
+## Install & init
+
+```bash
+cd Requirement_Agent
+pip install -e ".[dev]"
+
+# Scaffold config in the current directory
+req-agent init
+
+# Or into ~/.config/requirements-agent/
+req-agent init --global
+```
+
+### npm launcher (optional)
+
+The agent is Python. For `npx` / npm discoverability there is a thin wrapper in [`npm/`](npm/):
+
+```bash
+cd npm && npm install && npm link
+npx req-agent --help
+```
+
+Publish **PyPI** for the real package; publish **npm** only as the launcher (see [`npm/README.md`](npm/README.md)).
+
+One-shot release (same version on both registries; assumes you already ran `twine`/`npm` login):
+
+```bash
+./scripts/publish.sh           # current pyproject version
+./scripts/publish.sh 0.2.0     # bump + publish
+./scripts/publish.sh --dry-run # build only
+```
+
+`init` writes:
+
+| File | Purpose |
+|------|---------|
+| `requirements-agent.yaml` | Agent settings (context, prompts, export, active LLM profile) |
+| `llms.json` | Named LLM profiles (multiple models/providers) |
+| `.env.example` | Env var cheat-sheet for API keys |
+
+## Configure LLMs (`llms.json`)
+
+```json
+{
+  "default": "azure-gpt4o",
+  "models": {
+    "azure-gpt4o": {
+      "provider": "azure",
+      "model": "gpt-4o",
+      "deployment": "gpt-4o",
+      "endpoint": "https://YOUR.openai.azure.com/",
+      "api_key_env": "RA_AZURE_API_KEY",
+      "temperature": 0.3
+    },
+    "ollama-llama": {
+      "provider": "ollama",
+      "model": "llama3.2",
+      "endpoint": "http://127.0.0.1:11434/v1"
+    },
+    "mock": { "provider": "mock", "model": "mock" }
+  }
+}
+```
+
+List profiles: `req-agent models`
+
+## Configure agent (`requirements-agent.yaml`)
+
+```yaml
+llm:
+  models_file: ./llms.json
+  active: azure-gpt4o          # or override with --profile
+
+context:
+  providers: [folder]
+  folder_paths: [./docs]
+
+prompts:
+  personas:
+    architect:
+      system_extra: "Emphasize zero-trust boundaries."
+
+export:
+  default_format: both
+  out_dir: ./out
+```
+
+Examples: [`config.example.yaml`](config.example.yaml), [`llms.example.json`](llms.example.json). Legacy TOML still loads if present.
+
+Env overrides: `RA_CONFIG`, `RA_LLMS_JSON`, `RA_LLM_PROFILE`, `RA_LLM_PROVIDER`, `RA_LLM_MODEL`, `RA_API_KEY` / `OPENAI_API_KEY` / `RA_AZURE_*`.
+
+## CLI
+
+```bash
+# Offline dry-run
+req-agent run \
+  --title "Identity Hub SSO" \
+  --design-file ./brief.md \
+  --profile mock \
+  --format both \
+  --out ./out/identity-hub
+
+# Use a named profile from llms.json
+req-agent run --profile azure-gpt4o --title "..." --design-file brief.md
+
+# Interactive TUI
+req-agent tui --profile mock
+
+# Stages
+req-agent review --title "..." --design-file brief.md --out ./out
+req-agent score --title "..." --design-file brief.md --report ./out/01-spec-circuit-report.md
+req-agent refine --title "..." --design-file brief.md --report ./out/01-spec-circuit-report.md \
+  --message "Add measurable NFRs" --out ./out/01-refined.md
+req-agent consolidate --title "..." --design-file brief.md --report ./out/01-refined.md \
+  --out ./out/02-requirements-report.md
+req-agent export --report ./out/02-requirements-report.md --out ./out/report --format both
+```
+
+## Context injection
+
+| Provider | v1 |
+|----------|----|
+| **folder** | `--attach` / `--context-dir` / `folder_paths` |
+| **rag** | Stub (phase 2) |
+| **mcp** | Stub (phase 2) |
+
+## Personas
+
+Eight BMAD roles (vendored `bmad-core-spec`): orchestrator → analyst → pm → po → architect → dev → qa → sm.
+
+## TUIOS
+
+Optional: when running inside [tuios](https://github.com/Gaurav-Gosain/tuios), the agent calls `tuios set-agent-state`. Force with `RA_TUIOS_HOOK=1`.
+
+## vs Atlas Requirements Architect
+
+Same Spec Circuit brain and BMAD pack; Memory AI → local folder (future RAG/MCP); LLMs from `llms.json` instead of Mongo Settings. See [`REQUIREMENTS_ARCHITECT.md`](REQUIREMENTS_ARCHITECT.md).

requirements_agent-0.0.1/README.md

@@ -0,0 +1,143 @@
+# Requirements Agent
+
+Self-contained **Spec Circuit** requirements agent: multi-persona BMAD review → quality score → refine → BRD → Markdown/PDF.
+
+No Atlas / FastAPI / Mongo dependency. Configure LLMs in JSON, agent settings in YAML, run from CLI or Textual TUI.
+
+## Install & init
+
+```bash
+cd Requirement_Agent
+pip install -e ".[dev]"
+
+# Scaffold config in the current directory
+req-agent init
+
+# Or into ~/.config/requirements-agent/
+req-agent init --global
+```
+
+### npm launcher (optional)
+
+The agent is Python. For `npx` / npm discoverability there is a thin wrapper in [`npm/`](npm/):
+
+```bash
+cd npm && npm install && npm link
+npx req-agent --help
+```
+
+Publish **PyPI** for the real package; publish **npm** only as the launcher (see [`npm/README.md`](npm/README.md)).
+
+One-shot release (same version on both registries; assumes you already ran `twine`/`npm` login):
+
+```bash
+./scripts/publish.sh           # current pyproject version
+./scripts/publish.sh 0.2.0     # bump + publish
+./scripts/publish.sh --dry-run # build only
+```
+
+`init` writes:
+
+| File | Purpose |
+|------|---------|
+| `requirements-agent.yaml` | Agent settings (context, prompts, export, active LLM profile) |
+| `llms.json` | Named LLM profiles (multiple models/providers) |
+| `.env.example` | Env var cheat-sheet for API keys |
+
+## Configure LLMs (`llms.json`)
+
+```json
+{
+  "default": "azure-gpt4o",
+  "models": {
+    "azure-gpt4o": {
+      "provider": "azure",
+      "model": "gpt-4o",
+      "deployment": "gpt-4o",
+      "endpoint": "https://YOUR.openai.azure.com/",
+      "api_key_env": "RA_AZURE_API_KEY",
+      "temperature": 0.3
+    },
+    "ollama-llama": {
+      "provider": "ollama",
+      "model": "llama3.2",
+      "endpoint": "http://127.0.0.1:11434/v1"
+    },
+    "mock": { "provider": "mock", "model": "mock" }
+  }
+}
+```
+
+List profiles: `req-agent models`
+
+## Configure agent (`requirements-agent.yaml`)
+
+```yaml
+llm:
+  models_file: ./llms.json
+  active: azure-gpt4o          # or override with --profile
+
+context:
+  providers: [folder]
+  folder_paths: [./docs]
+
+prompts:
+  personas:
+    architect:
+      system_extra: "Emphasize zero-trust boundaries."
+
+export:
+  default_format: both
+  out_dir: ./out
+```
+
+Examples: [`config.example.yaml`](config.example.yaml), [`llms.example.json`](llms.example.json). Legacy TOML still loads if present.
+
+Env overrides: `RA_CONFIG`, `RA_LLMS_JSON`, `RA_LLM_PROFILE`, `RA_LLM_PROVIDER`, `RA_LLM_MODEL`, `RA_API_KEY` / `OPENAI_API_KEY` / `RA_AZURE_*`.
+
+## CLI
+
+```bash
+# Offline dry-run
+req-agent run \
+  --title "Identity Hub SSO" \
+  --design-file ./brief.md \
+  --profile mock \
+  --format both \
+  --out ./out/identity-hub
+
+# Use a named profile from llms.json
+req-agent run --profile azure-gpt4o --title "..." --design-file brief.md
+
+# Interactive TUI
+req-agent tui --profile mock
+
+# Stages
+req-agent review --title "..." --design-file brief.md --out ./out
+req-agent score --title "..." --design-file brief.md --report ./out/01-spec-circuit-report.md
+req-agent refine --title "..." --design-file brief.md --report ./out/01-spec-circuit-report.md \
+  --message "Add measurable NFRs" --out ./out/01-refined.md
+req-agent consolidate --title "..." --design-file brief.md --report ./out/01-refined.md \
+  --out ./out/02-requirements-report.md
+req-agent export --report ./out/02-requirements-report.md --out ./out/report --format both
+```
+
+## Context injection
+
+| Provider | v1 |
+|----------|----|
+| **folder** | `--attach` / `--context-dir` / `folder_paths` |
+| **rag** | Stub (phase 2) |
+| **mcp** | Stub (phase 2) |
+
+## Personas
+
+Eight BMAD roles (vendored `bmad-core-spec`): orchestrator → analyst → pm → po → architect → dev → qa → sm.
+
+## TUIOS
+
+Optional: when running inside [tuios](https://github.com/Gaurav-Gosain/tuios), the agent calls `tuios set-agent-state`. Force with `RA_TUIOS_HOOK=1`.
+
+## vs Atlas Requirements Architect
+
+Same Spec Circuit brain and BMAD pack; Memory AI → local folder (future RAG/MCP); LLMs from `llms.json` instead of Mongo Settings. See [`REQUIREMENTS_ARCHITECT.md`](REQUIREMENTS_ARCHITECT.md).

requirements_agent-0.0.1/pyproject.toml

@@ -0,0 +1,41 @@
+[build-system]
+requires = ["setuptools>=68", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "requirements-agent"
+version = "0.0.1"
+description = "Self-contained Spec Circuit Requirements Agent (CLI + Textual TUI)"
+readme = "README.md"
+requires-python = ">=3.9"
+license = { text = "MIT" }
+authors = [{ name = "CLI-Agents" }]
+dependencies = [
+  "typer>=0.12",
+  "rich>=13.7",
+  "textual>=0.80",
+  "openai>=1.40",
+  "httpx>=0.27",
+  "pyyaml>=6.0",
+  "tomli>=2.0; python_version < '3.11'",
+  "fpdf2>=2.7",
+  "markdown>=3.5",
+]
+
+[project.optional-dependencies]
+pdf = ["weasyprint>=62.0"]
+dev = ["pytest>=8.0", "pytest-asyncio>=0.23"]
+
+[project.scripts]
+req-agent = "requirements_agent.cli:app"
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.setuptools.package-data]
+requirements_agent = ["packs/**/*"]
+
+[tool.pytest.ini_options]
+asyncio_mode = "auto"
+testpaths = ["tests"]
+pythonpath = ["src"]

requirements_agent-0.0.1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

requirements_agent-0.0.1/src/requirements_agent/__init__.py

@@ -0,0 +1,3 @@
+"""Self-contained Spec Circuit Requirements Agent."""
+
+__version__ = "0.0.1"

requirements_agent-0.0.1/src/requirements_agent/agent.py

@@ -0,0 +1,143 @@
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Any, Callable, Dict, List, Optional
+
+from requirements_agent.config import AppConfig
+from requirements_agent.export.markdown import export_run_artifacts
+from requirements_agent.export.pdf import export_pdf
+from requirements_agent.llm.base import LlmClient
+from requirements_agent.llm.router import build_llm_client
+from requirements_agent.models import QualityScore, ReviewResult
+from requirements_agent.pipeline.brd import BRDConsolidator
+from requirements_agent.pipeline.quality import QualityScorer
+from requirements_agent.pipeline.refine import ReportRefiner
+from requirements_agent.pipeline.spec_circuit import SpecCircuitService
+from requirements_agent.tuios_hook import set_agent_state
+
+
+ProgressCallback = Callable[[Dict[str, Any]], None]
+
+
+class RequirementsAgent:
+    def __init__(self, config: AppConfig, llm: Optional[LlmClient] = None, *, extra_paths: Optional[List[str]] = None):
+        self.config = config
+        self.llm = llm or build_llm_client(config.llm)
+        self.extra_paths = list(extra_paths or [])
+        self.spec = SpecCircuitService.from_app_config(config, self.llm, extra_folder_paths=self.extra_paths)
+        self.quality = QualityScorer(self.llm, self.spec.context)
+        self.refiner = ReportRefiner(self.llm, self.spec.context)
+        self.brd = BRDConsolidator(self.llm, self.spec.context)
+
+    def review(self, title: str, design: str, *, on_progress: Optional[ProgressCallback] = None) -> ReviewResult:
+        set_agent_state("working", detail="Spec Circuit review")
+        try:
+            return self.spec.review(title, design, on_progress=on_progress)
+        finally:
+            set_agent_state("idle")
+
+    def score(
+        self,
+        title: str,
+        design: str,
+        report: str,
+        *,
+        refinement_context: str = "",
+    ) -> QualityScore:
+        set_agent_state("working", detail="Quality score")
+        try:
+            return self.quality.score(title, design, report, refinement_context=refinement_context)
+        finally:
+            set_agent_state("idle")
+
+    def refine(self, title: str, design: str, report: str, message: str, *, history=None) -> Dict[str, Any]:
+        set_agent_state("working", detail="Refine report")
+        try:
+            return self.refiner.refine(title, design, report, message, history=history)
+        finally:
+            set_agent_state("idle")
+
+    def questions(self, title: str, design: str, report: str) -> Dict[str, Any]:
+        set_agent_state("working", detail="Extract questions")
+        try:
+            return self.refiner.extract_open_questions(title, design, report)
+        finally:
+            set_agent_state("idle")
+
+    def resolve(self, title: str, design: str, report: str, question: str) -> Dict[str, Any]:
+        set_agent_state("working", detail="Suggest resolution")
+        try:
+            return self.refiner.suggest_question_resolution(title, design, report, question)
+        finally:
+            set_agent_state("idle")
+
+    def consolidate(self, title: str, design: str, report: str) -> Dict[str, Any]:
+        set_agent_state("working", detail="BRD consolidation")
+        try:
+            return self.brd.consolidate(title, design, report)
+        finally:
+            set_agent_state("idle")
+
+    def run(
+        self,
+        title: str,
+        design: str,
+        *,
+        out_dir: Optional[Path] = None,
+        export_format: Optional[str] = None,
+        consolidate: bool = True,
+        on_progress: Optional[ProgressCallback] = None,
+    ) -> Dict[str, Any]:
+        out = Path(out_dir or self.config.export.out_dir).expanduser()
+        fmt = (export_format or self.config.export.default_format or "md").lower()
+
+        review = self.review(title, design, on_progress=on_progress)
+        if not review.success:
+            return {"success": False, "error": review.error}
+
+        quality = self.score(title, design, review.report_markdown)
+        brd_md = None
+        if consolidate:
+            brd_result = self.consolidate(title, design, review.report_markdown)
+            if brd_result.get("success"):
+                brd_md = brd_result["report_markdown"]
+
+        quality_dict = {
+            "overall_score": quality.overall_score,
+            "dimensions": quality.dimensions,
+            "rationale": quality.rationale,
+            "top_gaps": quality.top_gaps,
+            "success": quality.success,
+            "error": quality.error,
+        }
+        paths = export_run_artifacts(
+            out,
+            spec_circuit=review.report_markdown,
+            brd=brd_md,
+            quality=quality_dict,
+            sections=review.sections,
+            meta={
+                "title": title,
+                "provider": self.config.llm.provider,
+                "model": self.config.llm.model,
+                "context": review.context.summary() if review.context else None,
+            },
+        )
+
+        pdf_paths = []
+        if fmt in ("pdf", "both"):
+            if review.report_markdown:
+                p = export_pdf(review.report_markdown, out / "01-spec-circuit-report.pdf", title=title)
+                pdf_paths.append(str(p))
+            if brd_md:
+                p = export_pdf(brd_md, out / "02-requirements-report.pdf", title=f"{title} — BRD")
+                pdf_paths.append(str(p))
+            paths["pdf"] = pdf_paths
+
+        return {
+            "success": True,
+            "paths": paths,
+            "quality": quality_dict,
+            "report_markdown": review.report_markdown,
+            "brd_markdown": brd_md,
+        }