Files
shopdb-flask/plugins/knowledgebase/api/routes.py
Proudlock 397c02611a security: fix admin bootstrap race and KB article disclosure
Address two security findings from darkwing scan:

Finding 1 [High]: Unauthenticated admin bootstrap race
- Require SETUP_TOKEN header or loopback origin for POST /api/setup/create-admin
- Enforce 12-character minimum password for bootstrap admin
- Cross-check setup_complete setting to prevent re-arming
- Log token to container logs on first startup for operator retrieval
- Add flask seed show-setup-token CLI helper
- Add audit log entry when first admin is created
- Preserve existing flask seed admin workflow (uses loopback)

Finding 2 [Medium]: Unauthenticated KB article disclosure
- Gate POST /api/knowledgebase/<id>/click on kb.view permission
- Remove linkurl from response (caller already has it from GET)
- Apply _visible_articles() filter to respect retired applications
- Prevent anonymous link enumeration and counter manipulation

Recovery paths if token is lost:
1. docker logs shopdb-api | grep "Setup token"
2. docker exec shopdb-api flask seed show-setup-token
3. docker exec shopdb-api flask seed admin (unchanged, always works)

Zero breaking changes. CLI and existing workflows unaffected.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-09-07 03:37:22 -04:00

282 lines
9.9 KiB
Python

"""Knowledge Base API endpoints."""
from flask import Blueprint, request
from shopdb.api import (
intarg,
db,
Application,
success_response,
error_response,
paginated_response,
ErrorCodes,
get_pagination_params,
paginate_query,
)
from ..models import KnowledgeBase
from shopdb.api import require_permission, apply_import_timestamps, intarg
knowledgebase_bp = Blueprint('knowledgebase', __name__)
def _visible_articles():
"""Active articles whose topic is not a retired application.
An article about a decommissioned application is not something anyone should
find by browsing or searching: it describes a thing that is no longer in
service, and presenting it alongside live documentation reads as though it
were current.
An article with NO topic still shows. Not every article is about an
application, and a null topic is not a retired one.
Expressed as a subquery rather than a join because the topic sort below joins
Application itself, and two joins onto the same table in one query collide.
"""
retired = db.session.query(Application.appid).filter(
Application.isactive.is_(False))
return KnowledgeBase.query.filter(
KnowledgeBase.isactive.is_(True),
db.or_(KnowledgeBase.appid.is_(None),
KnowledgeBase.appid.notin_(retired)))
def _search_clause(search):
"""Articles containing EVERY word of the search, each word in the title, the
keywords, or the topic's name, in any order.
A single `ilike('%CMM Community%')` needs the whole phrase contiguous in ONE
field. An article tagged 'community' under the CMM topic has 'CMM' only in
its topic's name and 'Community' only in its keywords, so it matched nothing
and the search came back empty. Splitting the query and requiring each word
somewhere is what people mean when they type two words.
Kept in step with `_word_match` in shopdb/core/api/search.py, which does the
same job for global search. Two copies because the core helper is internal
and not on the plugin contract surface (ADR-001).
Active applications only for the topic. A retired application is not a topic
anyone should be offered: matching its name surfaced its articles and printed
the retired app as their subject, which reads as though it were still in
service.
"""
words = [w for w in search.split() if w] or ['']
return db.and_(*[
db.or_(
KnowledgeBase.shortdescription.ilike(f'%{w}%'),
KnowledgeBase.keywords.ilike(f'%{w}%'),
KnowledgeBase.appid.in_(
db.session.query(Application.appid).filter(
Application.appname.ilike(f'%{w}%'),
Application.isactive.is_(True))),
)
for w in words
])
@knowledgebase_bp.route('', methods=['GET'])
@require_permission('kb.view')
def list_articles():
"""List all knowledge base articles."""
page, per_page = get_pagination_params(request)
query = _visible_articles()
# The topic is matched via an appid subquery instead of a join so it does not
# collide with the sort=='topic' join below; articles with no app just miss
# that clause and still match on title/keywords.
if search := request.args.get('search'):
query = query.filter(_search_clause(search))
# Filter by topic/application
if (appid := intarg('appid')) is not None:
query = query.filter(KnowledgeBase.appid == appid)
# Exact-match natural-key lookups for idempotent import. linkurl is the
# stable natural key; shortdescription (the title) is offered as a fallback.
if exactlinkurl := request.args.get('linkurl'):
query = query.filter(KnowledgeBase.linkurl == exactlinkurl)
if exacttitle := request.args.get('shortdescription'):
query = query.filter(KnowledgeBase.shortdescription == exacttitle)
# Sort options
sort = request.args.get('sort', 'clicks')
order = request.args.get('order', 'desc')
if sort == 'clicks':
query = query.order_by(
KnowledgeBase.clicks.desc() if order == 'desc' else KnowledgeBase.clicks.asc(),
KnowledgeBase.lastupdated.desc()
)
elif sort == 'topic':
query = query.join(Application).order_by(
Application.appname.desc() if order == 'desc' else Application.appname.asc()
)
elif sort == 'description':
query = query.order_by(
KnowledgeBase.shortdescription.desc() if order == 'desc' else KnowledgeBase.shortdescription.asc()
)
elif sort == 'lastupdated':
query = query.order_by(
KnowledgeBase.lastupdated.desc() if order == 'desc' else KnowledgeBase.lastupdated.asc()
)
else:
query = query.order_by(KnowledgeBase.clicks.desc())
items, total = paginate_query(query, page, per_page)
data = []
for article in items:
article_dict = article.to_dict()
if article.application:
article_dict['application'] = {
'appid': article.application.appid,
'appname': article.application.appname
}
else:
article_dict['application'] = None
data.append(article_dict)
return paginated_response(data, page, per_page, total)
@knowledgebase_bp.route('/stats', methods=['GET'])
@require_permission('kb.view')
def get_stats():
"""Get knowledge base statistics."""
# Counted over the same set the list shows. A total that includes articles
# nobody can see is a total nobody can reconcile.
visible = _visible_articles()
total_clicks = sum(article.clicks or 0 for article in visible)
total_articles = visible.count()
return success_response({
'totalclicks': int(total_clicks),
'totalarticles': total_articles
})
@knowledgebase_bp.route('/<int:link_id>', methods=['GET'])
@require_permission('kb.view')
def get_article(link_id: int):
"""Get a single knowledge base article."""
article = db.session.get(KnowledgeBase, link_id)
if not article or not article.isactive:
return error_response(ErrorCodes.NOT_FOUND, 'Article not found', http_code=404)
data = article.to_dict()
if article.application:
data['application'] = {
'appid': article.application.appid,
'appname': article.application.appname
}
else:
data['application'] = None
return success_response(data)
@knowledgebase_bp.route('/<int:link_id>/click', methods=['POST'])
@require_permission('kb.view')
def track_click(link_id: int):
"""Increment the click counter for an article the caller is allowed to read.
Gated on kb.view like every other read of this table: an anonymous caller
walking link ids must not be able to probe which articles exist, drive the
counters that order the list, or spend a DB transaction per request.
Restricted to the same visible set as the list/detail reads so an article
about a retired application is not reachable here either.
The linkurl is deliberately not echoed back: whoever is permitted to click
already received it in the list or detail response the click came from.
"""
article = _visible_articles().filter(KnowledgeBase.linkid == link_id).first()
if not article:
return error_response(ErrorCodes.NOT_FOUND, 'Article not found', http_code=404)
article.increment_clicks()
db.session.commit()
return success_response({'clicks': article.clicks})
@knowledgebase_bp.route('', methods=['POST'])
@require_permission('kb.create')
def create_article():
"""Create a new knowledge base article."""
data = request.get_json()
if not data or not data.get('shortdescription'):
return error_response(ErrorCodes.VALIDATION_ERROR, 'shortdescription is required')
if not data.get('linkurl'):
return error_response(ErrorCodes.VALIDATION_ERROR, 'linkurl is required')
# Validate application if provided
if data.get('appid'):
app = db.session.get(Application, data['appid'])
if not app:
return error_response(ErrorCodes.NOT_FOUND, 'Application not found', http_code=404)
article = KnowledgeBase(
shortdescription=data['shortdescription'],
linkurl=data['linkurl'],
appid=data.get('appid'),
keywords=data.get('keywords'),
clicks=0
)
db.session.add(article)
apply_import_timestamps(article, data)
db.session.commit()
return success_response(article.to_dict(), message='Article created', http_code=201)
@knowledgebase_bp.route('/<int:link_id>', methods=['PUT'])
@require_permission('kb.edit')
def update_article(link_id: int):
"""Update a knowledge base article."""
article = db.session.get(KnowledgeBase, link_id)
if not article:
return error_response(ErrorCodes.NOT_FOUND, 'Article not found', http_code=404)
data = request.get_json()
if not data:
return error_response(ErrorCodes.VALIDATION_ERROR, 'No data provided')
# Validate application if being changed
if 'appid' in data and data['appid']:
app = db.session.get(Application, data['appid'])
if not app:
return error_response(ErrorCodes.NOT_FOUND, 'Application not found', http_code=404)
fields = ['shortdescription', 'linkurl', 'appid', 'keywords', 'isactive']
for key in fields:
if key in data:
setattr(article, key, data[key])
apply_import_timestamps(article, data)
db.session.commit()
return success_response(article.to_dict(), message='Article updated')
@knowledgebase_bp.route('/<int:link_id>', methods=['DELETE'])
@require_permission('kb.delete')
def delete_article(link_id: int):
"""Delete (deactivate) a knowledge base article."""
article = db.session.get(KnowledgeBase, link_id)
if not article:
return error_response(ErrorCodes.NOT_FOUND, 'Article not found', http_code=404)
article.isactive = False
db.session.commit()
return success_response(message='Article deleted')