chore: remove orphan portal_scout code (SO no longer produces it) (#32)

* chore: remove orphan portal_scout code (SO no longer produces it)

* docs: remove portal_scout references from schema doc and README (SO no longer produces it)

Author: Gabrymi93

README.md

@@ -22,7 +22,6 @@ frontmatter o README manuali. Oggi consuma:
 |---|---|---|
 | `source-observatory` | `data/radar/radar_summary.json` | health complessivo delle fonti nel registry |
 | `source-observatory` | `data/catalog/catalog_signals.json` | drift/inventory per singola fonte |
-| `source-observatory` | `data/portal_scout/discovered_portals_summary.json` | nuovi portali PA scoperti da portal-scout |
 | `dataset-incubator` | `registry/pipeline_signals.json` | stato operativo dei dataset candidate |
 | `dataset-incubator` | `registry/clean_catalog.json` | dataset clean/queryable disponibili |
 

schemas/workspace_triage.md

@@ -39,8 +39,7 @@ Contiene stato incrociato di PR, issue, discussion, git e segnali operativi.
   "radar": { "generated_at": "...", "green": 20, "yellow": 2, "red": 0, "unhealthy": [] },
   "source_health": { "captured_at": "...", "alerts": [] },
   "pipeline_state": { "generated_at": "...", "actionable": [] },
-  "dataset_catalog": { "schema_version": "1", "updated_at": "...", "clean_ready": [] },
-  "portal_scout": { "generated_at": "...", "new_candidates": 0, "new_structured": [] }
+  "dataset_catalog": { "schema_version": "1", "updated_at": "...", "clean_ready": [] }
 }
 ```
 
@@ -66,7 +65,6 @@ Contiene stato incrociato di PR, issue, discussion, git e segnali operativi.
 | `source_health` | object \| null | Segnali drift/inventory da `catalog_signals.json` |
 | `pipeline_state` | object \| null | Stato candidati da `pipeline_signals.json` |
 | `dataset_catalog` | object \| null | Dataset clean-ready da `clean_catalog.json` |
-| `portal_scout` | object \| null | Portali scoperti da `discovered_portals_summary.json` |
 
 ## `git_state[repo]`
 
@@ -82,5 +80,5 @@ Contiene stato incrociato di PR, issue, discussion, git e segnali operativi.
 ## Valori null
 
 I campi `open_prs`, `open_issues`, `open_discussions` sono `null` in caso di errore fetch.
-Le sezioni `radar`, `source_health`, `pipeline_state`, `dataset_catalog`, `portal_scout`
+Le sezioni `radar`, `source_health`, `pipeline_state`, `dataset_catalog`
 sono `null` se l'artifact upstream non è disponibile o non è stato ancora prodotto.

src/agent_context_builder/render.py

@@ -12,10 +12,7 @@
 from .sources.di import DatasetIncubatorFetcher
 from .sources.so import SourceObservatoryFetcher
 from .signals import (
-    DICleanCatalog,
-    PortalScoutSummary,
     RadarSummary,
-    RepoSignals,
     SourceObservatorySignals,
 )
 from .triage import build_workspace_triage
@@ -65,8 +62,7 @@ def render_session_bootstrap(self) -> str:
         # ── SCOUTING ────────────────────────────────────────────────────
         radar = self._fetch_radar_summary()
         so = self._fetch_source_observatory_signals()
-        scout = self._fetch_portal_scout()
-        has_scouting = radar is not None or so is not None or scout is not None
+        has_scouting = radar is not None or so is not None
 
         if has_scouting:
             lines.append("## 🔍 SCOUTING")
@@ -107,19 +103,6 @@ def render_session_bootstrap(self) -> str:
                         f"({so.sources_checked} sources checked)"
                     )
 
-            # Portal scout
-            if scout is not None:
-                lines.append(
-                    f"**Portal Scout**: {scout.total_portals} portali · "
-                    f"{scout.new_candidates} nuovi candidati · "
-                    f"{scout.new_confirmed_protocol} strutturati"
-                )
-                if scout.new_structured:
-                    domains = [c.domain for c in scout.new_structured[:3]]
-                    extra = f" + {len(scout.new_structured) - 3} altri" if len(scout.new_structured) > 3 else ""
-                    lines.append(f"  · **Nuovi CKAN**: {', '.join(domains)}{extra}")
-            else:
-                lines.append("**Portal Scout**: unavailable")
             lines.append("")
 
         # ── INTAKE ────────────────────────────────────────────────────────
@@ -224,10 +207,6 @@ def _fetch_radar_summary(self) -> RadarSummary | None:
         return self._so_fetcher.fetch_radar_summary()
 
 
-    def _fetch_portal_scout(self) -> PortalScoutSummary | None:
-        return self._so_fetcher.fetch_portal_scout()
-
-
     def _fetch_source_observatory_signals(self) -> SourceObservatorySignals | None:
         return self._so_fetcher.fetch_catalog_signals()
 

src/agent_context_builder/signals.py

@@ -352,54 +352,3 @@ def parse_radar_summary(raw: str) -> RadarSummary:
         persistent_red=data.get("persistent_red", 0),
         sources=sources,
     )
-
-
-@dataclass
-class PortalCandidate:
-    """Single portal candidate from discovered_portals_summary.json."""
-
-    domain: str
-    protocol: str
-    probe_url: str
-
-
-@dataclass
-class PortalScoutSummary:
-    """Portal discovery summary from source-observatory discovered_portals_summary.json."""
-
-    generated_at: str
-    total_portals: int
-    new_candidates: int
-    new_confirmed_protocol: int
-    known_registry_seen: int
-    by_protocol: dict[str, int]
-    new_structured: list[PortalCandidate] = field(default_factory=list)
-    known_registry_healthcheck: list[PortalCandidate] = field(default_factory=list)
-
-
-def parse_portal_scout_summary(raw: str) -> PortalScoutSummary:
-    try:
-        data: dict[str, Any] = json.loads(raw)
-    except json.JSONDecodeError as exc:
-        raise ValueError(f"Invalid JSON: {exc}") from exc
-
-    def _parse_candidates(items: list[dict[str, Any]]) -> list[PortalCandidate]:
-        return [
-            PortalCandidate(
-                domain=c.get("domain", ""),
-                protocol=c.get("protocol", ""),
-                probe_url=c.get("probe_url", ""),
-            )
-            for c in items
-        ]
-
-    return PortalScoutSummary(
-        generated_at=data.get("generated_at", "unknown"),
-        total_portals=data.get("total_portals", 0),
-        new_candidates=data.get("new_candidates", 0),
-        new_confirmed_protocol=data.get("new_confirmed_protocol", 0),
-        known_registry_seen=data.get("known_registry_seen", 0),
-        by_protocol=data.get("by_protocol", {}),
-        new_structured=_parse_candidates(data.get("new_structured", [])),
-        known_registry_healthcheck=_parse_candidates(data.get("known_registry_healthcheck", [])),
-    )

src/agent_context_builder/sources/so.py

@@ -6,10 +6,8 @@
 
 from ..github import GitHubCollector
 from ..signals import (
-    PortalScoutSummary,
     RadarSummary,
     SourceObservatorySignals,
-    parse_portal_scout_summary,
     parse_radar_summary,
     parse_source_observatory_signals,
 )
@@ -21,7 +19,6 @@ class SourceObservatoryData:
 
     radar: RadarSummary | None
     catalog_signals: SourceObservatorySignals | None
-    portal_scout: PortalScoutSummary | None
 
 
 class SourceObservatoryFetcher:
@@ -31,14 +28,12 @@ def __init__(self, collector: GitHubCollector):
         self.collector = collector
         self._radar_cache: RadarSummary | None | object = _UNSET
         self._catalog_signals_cache: SourceObservatorySignals | None | object = _UNSET
-        self._portal_scout_cache: PortalScoutSummary | None | object = _UNSET
 
     def fetch(self) -> SourceObservatoryData:
         """Fetch all source-observatory artifacts."""
         return SourceObservatoryData(
             radar=self.fetch_radar_summary(),
             catalog_signals=self.fetch_catalog_signals(),
-            portal_scout=self.fetch_portal_scout(),
         )
 
     def fetch_radar_summary(self) -> RadarSummary | None:
@@ -71,23 +66,6 @@ def fetch_catalog_signals(self) -> SourceObservatorySignals | None:
         self._catalog_signals_cache = result
         return result
 
-    def fetch_portal_scout(self) -> PortalScoutSummary | None:
-        if self._portal_scout_cache is not _UNSET:
-            return self._portal_scout_cache  # type: ignore[return-value]
-        raw = self.collector.get_raw_file(
-            "source-observatory", "data/portal_scout/discovered_portals_summary.json"
-        )
-        if raw is None:
-            self._portal_scout_cache = None
-            return None
-        try:
-            result = parse_portal_scout_summary(raw)
-        except ValueError as exc:
-            self.collector.fetch_errors["source-observatory:portal_scout"] = str(exc)
-            result = None
-        self._portal_scout_cache = result
-        return result
-
 
 class _Unset:
     pass

src/agent_context_builder/triage.py

@@ -54,7 +54,6 @@ def build_workspace_triage(
         "source_health": _build_source_health_dict(so_fetcher, github_collector),
         "pipeline_state": _build_pipeline_state_dict(di_fetcher, github_collector),
         "dataset_catalog": _build_dataset_catalog_dict(di_fetcher, github_collector),
-        "portal_scout": _build_portal_scout_dict(so_fetcher),
     }
 
 
@@ -221,23 +220,6 @@ def _build_dataset_catalog_dict(
     }
 
 
-def _build_portal_scout_dict(fetcher: SourceObservatoryFetcher) -> dict[str, Any]:
-    scout = fetcher.fetch_portal_scout()
-    if scout is None:
-        return {"available": False}
-    return {
-        "available": True,
-        "generated_at": scout.generated_at,
-        "total_portals": scout.total_portals,
-        "new_candidates": scout.new_candidates,
-        "new_confirmed_protocol": scout.new_confirmed_protocol,
-        "by_protocol": scout.by_protocol,
-        "new_structured": [
-            {"domain": c.domain, "protocol": c.protocol} for c in scout.new_structured
-        ],
-    }
-
-
 def _collect_warnings(
     github_collector: GitHubCollector,
     prs: list[PR],

tests/test_render.py

@@ -398,14 +398,13 @@ def _raw_file_side_effect(repo, path, ref="main"):
     renderer.render_session_bootstrap()
     renderer.render_workspace_triage()
 
-    # Five distinct files fetched once each
-    assert gh.get_raw_file.call_count == 5
+    # Four distinct files fetched once each (portal_scout removed — SO no longer produces it)
+    assert gh.get_raw_file.call_count == 4
     paths_fetched = [call.args[1] for call in gh.get_raw_file.call_args_list]
     assert "data/radar/radar_summary.json" in paths_fetched
     assert "data/catalog/catalog_signals.json" in paths_fetched
     assert "registry/pipeline_signals.json" in paths_fetched
     assert "registry/clean_catalog.json" in paths_fetched
-    assert "data/portal_scout/discovered_portals_summary.json" in paths_fetched
 
 
 def test_render_topic_index():