Add async third-party vetting

Queue vetting on third_parties with PENDING, PROCESSING,
COMPLETED, and FAILED states. Expose enqueue and status through
GraphQL, MCP, CLI, and n8n, validate vet requests, tune the
worker via config, and poll the detail page while vetting runs.

Signed-off-by: Sacha Al Himdani <sacha@getprobo.com>

contrib/helm/charts/probo/templates/deployment.yaml

@@ -247,6 +247,36 @@ spec:
                   name: {{ include "probo.fullname" . }}
                   key: firecrawl-api-key
             {{- end }}
+            # Third-Party Vetter Agent
+            {{- if .Values.probo.thirdPartyVetter.provider }}
+            - name: AGENT_THIRD_PARTY_VETTER_PROVIDER
+              value: {{ .Values.probo.thirdPartyVetter.provider | quote }}
+            {{- end }}
+            {{- if .Values.probo.thirdPartyVetter.modelName }}
+            - name: AGENT_THIRD_PARTY_VETTER_MODEL_NAME
+              value: {{ .Values.probo.thirdPartyVetter.modelName | quote }}
+            {{- end }}
+            {{- if .Values.probo.thirdPartyVetter.temperature }}
+            - name: AGENT_THIRD_PARTY_VETTER_TEMPERATURE
+              value: {{ .Values.probo.thirdPartyVetter.temperature | quote }}
+            {{- end }}
+            {{- if .Values.probo.thirdPartyVetter.maxTokens }}
+            - name: AGENT_THIRD_PARTY_VETTER_MAX_TOKENS
+              value: {{ .Values.probo.thirdPartyVetter.maxTokens | quote }}
+            {{- end }}
+            # Third-Party Vetting Worker
+            {{- if .Values.probo.thirdPartyVettingWorker.interval }}
+            - name: THIRD_PARTY_VETTING_INTERVAL
+              value: {{ .Values.probo.thirdPartyVettingWorker.interval | quote }}
+            {{- end }}
+            {{- if .Values.probo.thirdPartyVettingWorker.maxConcurrency }}
+            - name: THIRD_PARTY_VETTING_MAX_CONCURRENCY
+              value: {{ .Values.probo.thirdPartyVettingWorker.maxConcurrency | quote }}
+            {{- end }}
+            {{- if .Values.probo.thirdPartyVettingWorker.staleAfter }}
+            - name: THIRD_PARTY_VETTING_STALE_AFTER
+              value: {{ .Values.probo.thirdPartyVettingWorker.staleAfter | quote }}
+            {{- end }}
             # Tracker Mapping Agent
             {{- if .Values.probo.trackerMapping.provider }}
             - name: AGENT_TRACKER_MAPPING_PROVIDER

contrib/helm/charts/probo/values-production.yaml.example

@@ -169,6 +169,19 @@ probo:
   # agentTools:
   #   firecrawlApiKey: "CHANGE_ME_FIRECRAWL_API_KEY"
 
+  # Third-party vetter agent (optional, AI-powered third-party vetting)
+  # thirdPartyVetter:
+  #   provider: "openai"
+  #   modelName: "gpt-4o"
+  #   temperature: "0.3"
+  #   maxTokens: "8192"
+
+  # Third-party vetting worker tuning (optional; seconds for interval/staleAfter).
+  # thirdPartyVettingWorker:
+  #   interval: 10
+  #   maxConcurrency: 1
+  #   staleAfter: 1500
+
   # Tracker mapping agent (optional, auto-links tracker patterns to vendors)
   # trackerMapping:
   #   provider: "openai"

contrib/helm/charts/probo/values.yaml

@@ -266,9 +266,24 @@ probo:
 
   # Agent tools (optional, shared across agents)
   agentTools:
-    # Firecrawl web search API key (used by tracker mapping and third-party assessor agents)
+    # Firecrawl web search API key (used by tracker mapping and third-party vetter agents)
     firecrawlApiKey: ""
 
+  # Third-party vetter agent (optional, requires openai.apiKey or anthropic key)
+  thirdPartyVetter:
+    provider: ""
+    modelName: ""
+    temperature: ""
+    maxTokens: ""
+
+  # Third-party vetting background worker tuning (optional). interval and
+  # staleAfter are in seconds. Keep concurrency modest to stay under LLM
+  # rate limits and the database connection pool.
+  thirdPartyVettingWorker:
+    interval: 10
+    maxConcurrency: 1
+    staleAfter: 1500
+
   # Tracker mapping agent (optional, requires openai.apiKey or anthropic key)
   trackerMapping:
     provider: ""