feat: configurable overview script

.github/workflows/refresh-profile.yml

@@ -37,7 +37,7 @@ jobs:
       - name: Collect metrics
         env:
           GITHUB_TOKEN: ${{ secrets.SCORE_BOT_PAT }}
-        run: uv run generate-repo-overview collect
+        run: uv run generate-repo-overview collect --org-config org_config.toml
 
       - name: Render overview (MD)
         run: uv run generate-repo-overview render-overview

AGENTS.md

@@ -6,7 +6,7 @@ Context file for AI coding assistants. See `docs/repo-overview-tool-design.md` f
 
 ```sh
 uv sync --all-groups --frozen                   # install deps
-uv run generate-repo-overview collect           # GitHub API → snapshot JSON
+uv run generate-repo-overview collect --org-config org_config.toml  # GitHub API → snapshot JSON
 uv run generate-repo-overview render-overview   # snapshot → profile/README.md
 uv run generate-repo-overview render-details    # snapshot → _site/ (index + per-repo pages)
 uv run pytest                                   # run tests
@@ -17,14 +17,18 @@ uv run basedpyright src/                        # type check
 ## Key files for website work
 
 ```
+org_config.toml                        — organization-specific settings (org name, tracked deps, workflow signals)
 src/generate_repo_overview/
+  org_config.py       — loads and validates org_config.toml
   metrics_html.py     — HTML renderer (index + per-repo detail pages)
   metrics_report.py   — shared helpers: grouping, version comparison, badges
-  models.py           — RepoEntry, RepoSnapshot, signal dataclasses
+  models.py           — RepoEntry, RepoSnapshot, TrackedDep, WorkflowSignal dataclasses
   cli.py              — render-details writes all pages from render_all_pages()
   constants.py        — default paths (DEFAULT_METRICS_HTML_OUTPUT = _site/)
 tests/
   test_cli_render.py  — render output tests
+  test_org_config.py  — org config loading/validation tests
+  test_repo_overview.py — collector and snapshot round-trip tests
 ```
 
 ## Website rendering notes

CLAUDE.md

@@ -1 +1 @@
-See [AGENTS.md](AGENTS.md) for project context, key files, and dev commands.
+@AGENTS.md

README.md

@@ -28,7 +28,7 @@ uv run generate-repo-overview render-details
 For a fresh GitHub pull before rendering, run:
 
 ```sh
-uv run generate-repo-overview collect
+uv run generate-repo-overview collect --org-config org_config.toml
 ```
 
 By default, `collect` now does a cache-aware refresh: it checks fast, high-level

docs/repo-overview-tool-design.md

@@ -11,15 +11,17 @@
 
 The tool is split into three layers:
 
-1. `collector/`
+1. `org_config.py`
+   - Loads organization-specific settings from `org_config.toml`: org name, repo include patterns, tracked Bazel deps, workflow signals, reference integration repo, and registry repo.
+2. `collector/`
    - Connects to GitHub.
    - Loads active repositories and custom properties.
-   - Derives content-based signals such as `has_ci`, `has_lint_config`, `has_coverage_config`, `bazel_version`, and `referenced_by_reference_integration`.
+   - Derives content-based signals such as `has_ci`, `has_lint_config`, `has_coverage_config`, `bazel_version`, `matched_workflow_signals`, `bazel_deps`, and `referenced_by_reference_integration`.
    - Writes and reads a local JSON snapshot cache.
-2. `profile_readme.py`, `metrics_report.py`, `metrics_html.py` (with `_html_index.py`, `_html_detail.py`, `_html_common.py`)
+3. `profile_readme.py`, `metrics_report.py`, `metrics_html.py` (with `_html_index.py`, `_html_detail.py`, `_html_common.py`)
    - Render different views (Markdown and HTML) from the same normalized data model.
    - Keep presentation decisions out of the collection layer.
-3. `cli.py`
+4. `cli.py`
    - Orchestrates cache-aware commands: `collect`, `render-overview`, and `render-details`.
 
 ## Data Model
@@ -32,6 +34,8 @@ The shared model lives in `models.py`.
   - organization name
   - generation timestamp
   - normalized repositories
+  - tracked Bazel dependency definitions (`tracked_deps`)
+  - workflow signal definitions (`workflow_signals`)
 
 The snapshot is intentionally renderer-agnostic. It stores neutral values such as booleans and plain strings rather than Markdown-specific markers.
 
@@ -78,8 +82,9 @@ uv run generate-repo-overview <command>
 
 Built-in commands:
 
-- `collect`
+- `collect --org-config org_config.toml`
   - Sync the cached snapshot from GitHub and write it to disk.
+  - Requires `--org-config` pointing to a TOML file with organization-specific settings.
   - Use `--deep` to force a full refresh for every repository instead of reusing cached signals for unchanged ones.
 - `render-overview`
   - Render the profile README from an existing snapshot.

org_config.toml

@@ -0,0 +1,40 @@
+# Organization configuration for generate-repo-overview.
+#
+# This file defines which GitHub organization to scan and how to detect
+# repository signals.  The tool requires this file via --org-config.
+#
+# For a different organization, copy this file and adjust the values.
+
+org_name = "eclipse-score"
+
+# Optional fnmatch glob patterns to limit which repositories are included.
+# When empty (or omitted), all non-archived repositories in the org are included.
+# Example: repo_include_patterns = ["my-prefix-*", "another-repo"]
+# repo_include_patterns = []
+
+[signals]
+
+# Repository (org/repo) that serves as the reference integration project.
+# Its MODULE.bazel is parsed to determine which other repositories are
+# referenced as Bazel dependencies.
+# Leave empty if there is no reference integration repository.
+reference_integration_repo = "eclipse-score/reference_integration"
+
+# Full "org/repo" path to the Bazel registry repository.
+# Leave empty if there is no Bazel registry.
+registry_repo = "eclipse-score/bazel_registry"
+
+# Tracked Bazel dependencies.  Each entry maps a source repository (org/repo)
+# to its Bazel module name.  The tool looks up each module_name in every
+# repository's MODULE.bazel to report dependency versions.
+[[signals.tracked_deps]]
+repo = "eclipse-score/docs-as-code"
+module_name = "score_docs_as_code"
+
+# Named workflow signals.  Each entry has a human-readable label shown in the
+# metrics output, and a reference string to search for inside
+# .github/workflows/*.yml files.  A repository is flagged for a signal when
+# any of its workflow files contain the reference string.
+[[signals.workflow_signals]]
+label = "Daily Workflow"
+reference = "eclipse-score/cicd-workflows/.github/workflows/daily.yml@"

src/generate_repo_overview/README.md

@@ -38,8 +38,10 @@ This document explains the package structure and cache behavior. It intentionall
   - Renders the main HTML metrics dashboard (tabs, filters, sortable columns).
 - `_html_detail.py`
   - Renders per-repository HTML detail pages.
+- `org_config.py`
+  - Loads `org_config.toml`: org name, repo include patterns, tracked Bazel deps, workflow signals.
 - `constants.py`
-  - Centralizes default org, cache, and output paths.
+  - Centralizes default cache and output paths.
 - `console.py`
   - Keeps status output formatting in one place.
 
@@ -70,6 +72,8 @@ That file stores a serialized `RepoSnapshot` containing:
 - organization name
 - generation timestamp
 - all normalized repositories
+- tracked Bazel dependency definitions (`tracked_deps`)
+- workflow signal definitions (`workflow_signals`)
 
 The cache loader only accepts the current schema version. If the snapshot schema does not match, the cache is treated as unusable and collection falls back to a fresh GitHub fetch.
 
@@ -106,14 +110,13 @@ For each repository, the snapshot currently stores:
   - `is_bazel_repo`
   - `bazel_version`
   - `codeowners`
-  - `docs_as_code_version`
   - `referenced_by_reference_integration`
   - `has_lint_config`
   - `has_gitlint_config`
   - `has_pyproject_toml`
   - `has_pre_commit_config`
   - `has_ci`
-  - `uses_cicd_daily_workflow`
+  - `matched_workflow_signals`
   - `has_coverage_config`
   - `top_languages`
   - `bazel_deps`

src/generate_repo_overview/_html_common.py

@@ -81,9 +81,15 @@ def version_badge(
     version: str | None,
     max_bazel: tuple[int, ...] | None,
     *,
-    latest_dac: str | None,
+    latest_dep_version: str | None,
     is_bazel: bool,
 ) -> str:
+    """Render a colored version badge span.
+
+    Bazel versions are green when equal to *max_bazel*, red otherwise.
+    Dep versions compare against *latest_dep_version*: green if equal,
+    yellow if same major.minor, red if older, muted if unknown.
+    """
     if version is None or not version.strip():
         return '<span class="badge muted">—</span>'
 
@@ -95,9 +101,9 @@ def version_badge(
             return f'<span class="badge green">{e(cleaned)}</span>'
         return f'<span class="badge red">{e(cleaned)}</span>'
 
-    if latest_dac is None:
+    if latest_dep_version is None:
         return f'<span class="badge muted">{e(cleaned)}</span>'
-    latest_cleaned = latest_dac.strip()
+    latest_cleaned = latest_dep_version.strip()
     if cleaned == latest_cleaned:
         return f'<span class="badge green">{e(cleaned)}</span>'
     if parsed is not None:

src/generate_repo_overview/_html_detail.py

@@ -3,18 +3,19 @@
 from typing import TYPE_CHECKING
 
 from ._html_common import BAZEL_ICON, CSS, GITHUB_ICON, e, language_badge, version_badge
+from .metrics_report import tracked_dep_label
 
 if TYPE_CHECKING:
     from .models import RepoEntry, RepoSnapshot
 
 
 def render_detail_page(
     entry: RepoEntry,
-    org_name: str,
     snapshot: RepoSnapshot,
     max_bazel: tuple[int, ...] | None,
-    latest_dac: str | None,
+    latest_dep_versions: dict[str, str | None],
 ) -> str:
+    org_name = snapshot.org_name
     return (
         "<!DOCTYPE html>\n"
         '<html lang="en">\n<head>\n'
@@ -27,9 +28,9 @@ def render_detail_page(
         + _render_stat_grid(entry)
         + _render_release_section(entry)
         + _render_dep_diff_section(entry)
-        + _render_tooling_section(entry)
+        + _render_tooling_section(entry, snapshot)
         + _render_ownership_section(entry)
-        + _render_versions_section(entry, max_bazel, latest_dac)
+        + _render_versions_section(entry, snapshot, max_bazel, latest_dep_versions)
         + _render_footer(snapshot)
         + "</body>\n</html>\n"
     )
@@ -231,18 +232,21 @@ def _dep_status_badge(status: str, css_class: str) -> str:
     return f'<span class="badge {css_class}">{e(status)}</span>'
 
 
-def _render_tooling_section(entry: RepoEntry) -> str:
+def _render_tooling_section(entry: RepoEntry, snapshot: RepoSnapshot) -> str:
     c = entry.content
-    signals = [
+    signals: list[tuple[bool, str]] = [
         (c.has_ci, "GitHub Actions (CI)"),
-        (c.uses_cicd_daily_workflow, "Daily Workflow"),
+    ]
+    for label in (s.label for s in snapshot.workflow_signals):
+        signals.append((label in c.matched_workflow_signals, label))
+    signals.extend([
         (c.has_lint_config, "Lint Config"),
         (c.has_gitlint_config, "Gitlint"),
         (c.has_pre_commit_config, "Pre-commit"),
         (c.has_pyproject_toml, "pyproject.toml"),
         (c.has_coverage_config, "Coverage Config"),
         (c.is_bazel_repo, "Bazel Repo"),
-    ]
+    ])
 
     items = "\n".join(
         f'    <div class="signal-item">'
@@ -288,26 +292,31 @@ def _render_ownership_section(entry: RepoEntry) -> str:
 
 def _render_versions_section(
     entry: RepoEntry,
+    snapshot: RepoSnapshot,
     max_bazel: tuple[int, ...] | None,
-    latest_dac: str | None,
+    latest_dep_versions: dict[str, str | None],
 ) -> str:
+    from .models import lookup_bazel_dep_version
+
     items: list[str] = []
 
     bazel_badge = version_badge(
-        entry.content.bazel_version, max_bazel, latest_dac=None, is_bazel=True
+        entry.content.bazel_version, max_bazel, latest_dep_version=None, is_bazel=True
     )
     items.append(
         f'<div class="info-item">'
         f'<div class="info-label">Bazel Version</div>{bazel_badge}</div>'
     )
 
-    dac_badge = version_badge(
-        entry.content.docs_as_code_version, None, latest_dac=latest_dac, is_bazel=False
-    )
-    items.append(
-        f'<div class="info-item">'
-        f'<div class="info-label">Docs-As-Code Version</div>{dac_badge}</div>'
-    )
+    for dep in snapshot.tracked_deps:
+        dep_label = tracked_dep_label(dep)
+        dep_ver = lookup_bazel_dep_version(entry.content.bazel_deps, dep.module_name)
+        latest_ver = latest_dep_versions.get(dep.module_name)
+        badge = version_badge(dep_ver, None, latest_dep_version=latest_ver, is_bazel=False)
+        items.append(
+            f'<div class="info-item">'
+            f'<div class="info-label">{e(dep_label)} Version</div>{badge}</div>'
+        )
 
     refint = (
         '<span class="badge green">yes</span>'