~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.
Python academiaanthropicarxivbravedeep-researchencryptionhome-automationhomeserverlocallocal-deep-researchlocal-llmmistralollamaopenaipubmedresearchresearch-toolretrieval-augmented-generationsearxngself-hosted
fix(ci): standardize server-readiness probes (curl -fsS /api/v1/health) + make responsive-ui cleanup real (#4794)
LearningCircuitbc7b3a7c4b7f
diff --git a/.github/workflows/docker-multiarch-test.yml b/.github/workflows/docker-multiarch-test.ymlindex dc3305669b..19d556e582 100644--- a/.github/workflows/docker-multiarch-test.yml+++ b/.github/workflows/docker-multiarch-test.yml@@ -91,7 +91,7 @@ jobs:TIMEOUT_SECONDS=60echo "Waiting for server to start (timeout: ${TIMEOUT_SECONDS}s)..."for i in $(seq 1 ${TIMEOUT_SECONDS}); do- if curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000 2>/dev/null; then+ if curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health 2>/dev/null; thenecho "✅ Server started successfully after ${i} seconds on ${{ matrix.platform }}!"breakfidiff --git a/.github/workflows/nuclei.yml b/.github/workflows/nuclei.ymlindex 5ebe37a446..12f63157fb 100644--- a/.github/workflows/nuclei.yml+++ b/.github/workflows/nuclei.yml@@ -93,7 +93,7 @@ jobs:# --connect-timeout/--max-time bound TCP and total request time so a# hung connection fails fast instead of eating the job timeout.for _ in {1..90}; do- if curl -s --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; then+ if curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; thenecho "Server started successfully"breakfi@@ -101,7 +101,7 @@ jobs:done# Verify server is running- if ! curl -s --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; then+ if ! curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; thenecho "Server failed to start"cat server.logexit 1diff --git a/.github/workflows/owasp-zap-scan.yml b/.github/workflows/owasp-zap-scan.ymlindex 2171c77a0b..e243a71490 100644--- a/.github/workflows/owasp-zap-scan.yml+++ b/.github/workflows/owasp-zap-scan.yml@@ -64,7 +64,7 @@ jobs:# --connect-timeout/--max-time bound TCP and total request time so a# hung connection fails fast instead of eating the job timeout.for _ in {1..90}; do- if curl -s --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; then+ if curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; thenecho "Server started successfully"breakfi@@ -72,7 +72,7 @@ jobs:done# Verify server is running- if ! curl -s --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; then+ if ! curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; thenecho "Server failed to start"cat server.logexit 1@@ -161,7 +161,7 @@ jobs:# --connect-timeout/--max-time bound TCP and total request time so a# hung connection fails fast instead of eating the job timeout.for _ in {1..90}; do- if curl -s --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; then+ if curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; thenecho "Server started successfully"breakfi@@ -169,7 +169,7 @@ jobs:done# Verify server is running- if ! curl -s --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; then+ if ! curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; thenecho "Server failed to start"cat server.logexit 1diff --git a/.github/workflows/playwright-webkit-tests.yml b/.github/workflows/playwright-webkit-tests.ymlindex e57a7dc1b9..1a28844d90 100644--- a/.github/workflows/playwright-webkit-tests.yml+++ b/.github/workflows/playwright-webkit-tests.yml@@ -106,7 +106,7 @@ jobs:# --connect-timeout/--max-time bound TCP and total request time so a# hung connection fails fast instead of eating the job timeout.for i in {1..30}; do- if curl -s --connect-timeout 2 --max-time 5 http://127.0.0.1:5000 > /dev/null; then+ if curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; thenecho "Server is ready"breakfi@@ -115,7 +115,7 @@ jobs:done# Verify server is running- if ! curl -s --connect-timeout 2 --max-time 5 http://127.0.0.1:5000 > /dev/null; then+ if ! curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; thenecho "Server failed to start"exit 1fi@@ -276,7 +276,7 @@ jobs:# --connect-timeout/--max-time bound TCP and total request time so a# hung connection fails fast instead of eating the job timeout.for i in {1..30}; do- if curl -s --connect-timeout 2 --max-time 5 http://127.0.0.1:5000 > /dev/null; then+ if curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; thenecho "Server is ready"breakfi@@ -285,7 +285,7 @@ jobs:done# Verify server is running- if ! curl -s --connect-timeout 2 --max-time 5 http://127.0.0.1:5000 > /dev/null; then+ if ! curl -fsS --connect-timeout 2 --max-time 5 http://127.0.0.1:5000/api/v1/health > /dev/null; thenecho "Server failed to start"exit 1fidiff --git a/.github/workflows/puppeteer-e2e-tests.yml b/.github/workflows/puppeteer-e2e-tests.ymlindex 7dcdb69ad2..10cfa0e504 100644--- a/.github/workflows/puppeteer-e2e-tests.yml+++ b/.github/workflows/puppeteer-e2e-tests.yml@@ -160,7 +160,7 @@ jobs:echo "Waiting for server to start..."for i in {1..60}; do- if curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/ 2>/dev/null; then+ if curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health 2>/dev/null; thenecho "Server is ready after $i seconds!"breakfi@@ -174,7 +174,7 @@ jobs:done# Final check if server is ready- if ! curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/ 2>/dev/null; then+ if ! curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health 2>/dev/null; thenecho "Server failed to start. Logs:"cat /tmp/ldr_server.logexit 1diff --git a/.github/workflows/responsive-ui-tests-enhanced.yml b/.github/workflows/responsive-ui-tests-enhanced.ymlindex 34fa0069bf..ba8054b6ff 100644--- a/.github/workflows/responsive-ui-tests-enhanced.yml+++ b/.github/workflows/responsive-ui-tests-enhanced.yml@@ -144,6 +144,11 @@ jobs:pdm run python -m local_deep_research.web.app > server.log 2>&1 &SERVER_PID=$!echo "Server PID: $SERVER_PID"+ # Persist the PID for the "Stop application server" step. That step+ # runs at the workspace root (no working-directory) while we are in+ # src/ here, so write an absolute path — otherwise the pidfile never+ # lines up and cleanup is a silent no-op.+ echo "$SERVER_PID" > "$GITHUB_WORKSPACE/server.pid"# Wait for server to startSERVER_READY=false@@ -292,6 +297,11 @@ jobs:kill "$(cat server.pid)" || truerm server.pidfi+ # `pdm run` stays alive as a wrapper, so the pidfile holds the+ # wrapper PID and killing it can orphan the actual python server.+ # Add the same pkill fallback the webkit workflow uses so the+ # python process (and port 5000) is actually released.+ pkill -f "python -m local_deep_research.web.app" || truepost-results:needs: ui-testsdiff --git a/.github/workflows/security-headers-validation.yml b/.github/workflows/security-headers-validation.ymlindex 593ce43d78..ef1e3ad023 100644--- a/.github/workflows/security-headers-validation.yml+++ b/.github/workflows/security-headers-validation.yml@@ -88,7 +88,7 @@ jobs:- name: Wait for application to be readyrun: |- timeout 60 bash -c 'until curl -s --max-time 5 http://localhost:5000/ > /dev/null; do sleep 2; done'+ timeout 60 bash -c 'until curl -fsS --connect-timeout 2 --max-time 5 http://localhost:5000/api/v1/health > /dev/null; do sleep 2; done'- name: Validate security headersrun: |