Sprint 4D: AGENTS.md parser

purpose_agent/protocols/agents_md.py

@@ -0,0 +1,67 @@
+"""
+agents_md.py — Parser for repo-local AGENTS.md instruction files.
+
+Precedence: system_policy > developer_config > AGENTS.md > team_purpose > skill_cards > user_task
+"""
+from __future__ import annotations
+import logging
+from dataclasses import dataclass, field
+from pathlib import Path
+from typing import Any
+
+logger = logging.getLogger(__name__)
+
+@dataclass
+class AgentsConfig:
+    instructions: list[str] = field(default_factory=list)
+    capabilities: dict[str, str] = field(default_factory=dict)
+    constraints: list[str] = field(default_factory=list)
+    metadata: dict[str, Any] = field(default_factory=dict)
+    source_path: str = ""
+    def to_prompt_section(self) -> str:
+        parts = []
+        if self.instructions:
+            parts.append("## Repository Instructions (AGENTS.md)")
+            for i in self.instructions: parts.append(f"- {i}")
+        if self.constraints:
+            parts.append("\n## Constraints")
+            for c in self.constraints: parts.append(f"- {c}")
+        return "\n".join(parts)
+    @property
+    def has_content(self) -> bool:
+        return bool(self.instructions or self.capabilities or self.constraints)
+
+def parse_agents_md(content: str, source_path: str = "") -> AgentsConfig:
+    config = AgentsConfig(source_path=source_path)
+    if content.startswith("---"):
+        end = content.find("---", 3)
+        if end > 0: content = content[end+3:]
+    current = None
+    for line in content.split("\n"):
+        line = line.strip()
+        if line.startswith("## ") or line.startswith("# "):
+            h = line.lstrip("#").strip().lower()
+            if "instruction" in h: current = "instructions"
+            elif "capabilit" in h: current = "capabilities"
+            elif "constraint" in h: current = "constraints"
+            else: current = None
+            continue
+        if not line or line.startswith("<!--"): continue
+        item = line.lstrip("- ").lstrip("* ").strip()
+        if not item: continue
+        if current == "instructions": config.instructions.append(item)
+        elif current == "capabilities":
+            if ":" in item:
+                n, d = item.split(":", 1); config.capabilities[n.strip()] = d.strip()
+            else: config.capabilities[item] = ""
+        elif current == "constraints": config.constraints.append(item)
+    return config
+
+def load_agents_md(path: str = "AGENTS.md") -> AgentsConfig | None:
+    for p in [Path(path), Path(".github/AGENTS.md"), Path("docs/AGENTS.md")]:
+        if p.exists():
+            try:
+                config = parse_agents_md(p.read_text(), str(p))
+                if config.has_content: return config
+            except: pass
+    return None