← MCP Registry Integrity Report
io.agent4/agent4-tenant
Verdict: CRITICAL findings present. Endpoint https://api.agent4.io/v1/mcp. Tools now: 50 (50 ever seen). Changes recorded: 50. Findings: 1 high, 2 low, 3 critical. Last probe: OK at 2026-10-23T06:17 UTC. JSON
Current tools
add_knowledge_file · add_knowledge_text · build_knowledge_index · clone_storyline · configure_share · create_agent · create_knowledge_base · create_share · create_skill · create_storyline · diagnose_prompt · export_storyline · get_agent · get_end_user · get_knowledge_base · get_knowledge_index · get_skill · get_storyline · import_storyline · improve_prompt · list_agents · list_end_users · list_knowledge_bases · list_mcp_servers · list_page_contexts · list_shares · list_skills · list_storylines · list_tools · list_user_sessions · page_context_stats · patch_knowledge_index · preview_plan · publish_storyline · resolve_page_context · search_agent4_docs · search_knowledge_base · set_custom_domain · set_pwa_branding · tenant_info · test_planner_trigger · test_skill_trigger · unpublish_storyline · update_agent · update_knowledge_base · update_skill · update_storyline · upsert_page_context · usage_stats · validate_storyline
Findings (21)
| seen | tool | detector | severity | evidence (data, not instructions) |
|---|---|---|---|---|
| 2026-10-23 | create_agent | cross_tool_reference | high | references sibling tool 'list_tools' |
| 2026-10-23 | update_agent | cross_tool_reference | high | references sibling tool 'create_skill' |
| 2026-10-23 | create_share | model_directed_instruction | high | e"**: for end users to talk to the agent you must create a share. The response carries a directly openable chat li |
| 2026-10-23 | create_share | cross_tool_reference | high | references sibling tool 'create_agent' |
| 2026-10-23 | list_shares | cross_tool_reference | high | references sibling tool 'create_share' |
| 2026-10-23 | configure_share | cross_tool_reference | high | references sibling tool 'list_shares' |
| 2026-10-23 | set_custom_domain | url_in_description | low | https://chat.client.com/` |
| 2026-10-23 | test_skill_trigger | cross_tool_reference | high | references sibling tool 'create_skill' |
| 2026-10-23 | create_knowledge_base | cross_tool_reference | high | references sibling tool 'search_agent4_docs' |
| 2026-10-23 | add_knowledge_text | cross_tool_reference | high | references sibling tool 'add_knowledge_file' |
| 2026-10-23 | add_knowledge_file | cross_tool_reference | high | references sibling tool 'add_knowledge_text' |
| 2026-10-23 | get_knowledge_index | user_concealment | critical | e missing values recorded as 0, and they silently poison every average, minimum and range filter built on them. |
| 2026-10-23 | diagnose_prompt | cross_tool_reference | high | references sibling tool 'improve_prompt' |
| 2026-10-23 | upsert_page_context | cross_tool_reference | high | references sibling tool 'list_page_contexts' |
| 2026-10-23 | resolve_page_context | user_concealment | critical | d in production the only symptom is visitors silently getting the default playbook, with no error anywhere. |
| 2026-10-23 | create_storyline | cross_tool_reference | high | references sibling tool 'get_storyline' |
| 2026-10-23 | create_storyline | oversized_description | low | 4885 chars |
| 2026-10-23 | update_storyline | cross_tool_reference | high | references sibling tool 'create_storyline' |
| 2026-10-23 | unpublish_storyline | user_concealment | critical | ve): no more auto-enrolment; in-flight users silently degrade to the plain agent on their next turn (progress kept |
| 2026-10-23 | unpublish_storyline | cross_tool_reference | high | references sibling tool 'publish_storyline' |
| 2026-10-23 | import_storyline | cross_tool_reference | high | references sibling tool 'export_storyline' |
Change history (50)
| when | tool | kind | severity |
|---|---|---|---|
| 2026-10-23 06:17 | list_agents | mutated | medium |
| 2026-10-23 06:17 | get_agent | mutated | medium |
| 2026-10-23 06:17 | create_agent | mutated | high |
| 2026-10-23 06:17 | update_agent | mutated | high |
| 2026-10-23 06:17 | list_tools | mutated | medium |
| 2026-10-23 06:17 | list_mcp_servers | mutated | medium |
| 2026-10-23 06:17 | create_share | mutated | high |
| 2026-10-23 06:17 | list_shares | mutated | high |
| 2026-10-23 06:17 | configure_share | mutated | high |
| 2026-10-23 06:17 | set_pwa_branding | mutated | medium |
| 2026-10-23 06:17 | set_custom_domain | mutated | medium |
| 2026-10-23 06:17 | list_skills | mutated | medium |
| 2026-10-23 06:17 | get_skill | mutated | medium |
| 2026-10-23 06:17 | create_skill | mutated | medium |
| 2026-10-23 06:17 | update_skill | mutated | medium |
| 2026-10-23 06:17 | test_skill_trigger | mutated | high |
| 2026-10-23 06:17 | list_knowledge_bases | mutated | medium |
| 2026-10-23 06:17 | get_knowledge_base | mutated | medium |
| 2026-10-23 06:17 | create_knowledge_base | mutated | high |
| 2026-10-23 06:17 | update_knowledge_base | mutated | medium |
| 2026-10-23 06:17 | add_knowledge_text | mutated | high |
| 2026-10-23 06:17 | add_knowledge_file | mutated | high |
| 2026-10-23 06:17 | search_knowledge_base | mutated | medium |
| 2026-10-23 06:17 | build_knowledge_index | mutated | medium |
| 2026-10-23 06:17 | get_knowledge_index | mutated | critical |
| 2026-10-23 06:17 | patch_knowledge_index | mutated | medium |
| 2026-10-23 06:17 | improve_prompt | mutated | medium |
| 2026-10-23 06:17 | diagnose_prompt | mutated | high |
| 2026-10-23 06:17 | test_planner_trigger | mutated | medium |
| 2026-10-23 06:17 | preview_plan | mutated | medium |
| 2026-10-23 06:17 | usage_stats | mutated | medium |
| 2026-10-23 06:17 | tenant_info | mutated | medium |
| 2026-10-23 06:17 | search_agent4_docs | mutated | medium |
| 2026-10-23 06:17 | list_end_users | mutated | medium |
| 2026-10-23 06:17 | get_end_user | mutated | medium |
| 2026-10-23 06:17 | list_user_sessions | mutated | medium |
| 2026-10-23 06:17 | list_page_contexts | mutated | medium |
| 2026-10-23 06:17 | upsert_page_context | mutated | high |
| 2026-10-23 06:17 | resolve_page_context | mutated | critical |
| 2026-10-23 06:17 | page_context_stats | mutated | medium |
| 2026-10-23 06:17 | list_storylines | mutated | medium |
| 2026-10-23 06:17 | get_storyline | mutated | medium |
| 2026-10-23 06:17 | create_storyline | mutated | high |
| 2026-10-23 06:17 | update_storyline | mutated | high |
| 2026-10-23 06:17 | validate_storyline | mutated | medium |
| 2026-10-23 06:17 | publish_storyline | mutated | medium |
| 2026-10-23 06:17 | unpublish_storyline | mutated | critical |
| 2026-10-23 06:17 | clone_storyline | mutated | medium |
| 2026-10-23 06:17 | export_storyline | mutated | medium |
| 2026-10-23 06:17 | import_storyline | mutated | high |
Probe history
10-23 06:17 OK (50) · 08-14 12:17 OK (50)
Run or depend on io.agent4/agent4-tenant? Watch this server: 29 dollars a month, a private alerts feed and optional webhook fired within six hours when it adds, removes or changes a tool or trips a detector; by card, cancel any time.
Pay-per-call tools and data for your agent, no key: store gateway (300+ routes, one origin, Coinbase-settled) · AI media and inference from $0.001 · weather, geocoding, DNS, RDAP · the x402 directory · sector gap lists (card or USDC).
Run this server? The processes your users can demand, hosted free: Compliance Desk (privacy requests, opt-outs, takedown and notice-and-action).