New Your team’s decisions, in one playbook every coding agent works from. Never answer your agent twice
// SCAN REPORT

ProAgentStore

243 tools. 124 can modify or destroy data without limits.

24 destructive tools with no built-in limits. Policy required.

Last updated:

124 can modify or destroy data
119 read-only
243 tools total

Community server · catalogue entry checked 01/10/2026

How to control ProAgentStore ↓

TOOL MAP

What ProAgentStore exposes to your agents

Read (119) Write / Execute (99) Destructive / Financial (24)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS
Critical Risk

The most dangerous ProAgentStore tools

Financial · High cancel_instance This tool cancels a subscription, which is a financial action (terminating a financial commitment/obligation). Financial · High set_budget_limits Modifies financial spend limits controlling AI budget, directly affecting financial obligations. Destructive · High cancel_instance_task Cancelling a running task is an irreversible action — the in-progress work is terminated and cannot be resumed. Destructive · High clear_finished_tasks 'Clear finished tasks' likely irreversibly removes completed task records from the system.

124 of ProAgentStore's 243 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control ProAgentStore

PolicyLayer is an MCP gateway — it sits between your AI agents and ProAgentStore, and nothing reaches the server without passing your rules. These are the rules we recommend:

Block financial tools by default
{
  "cancel_instance": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Requires human approval."
      }
    ]
  }
}

Financial tools should be explicitly enabled per use case, not open by default.

Deny destructive operations
{
  "cancel_instance_task": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations
{
  "add_instance_knowledge": {
    "limits": [
      {
        "counter": "add_instance_knowledge_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "account_activity": {
    "limits": [
      {
        "counter": "account_activity_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

  1. Create a free account and register ProAgentStore — nothing to install.
  2. Add these rules — paste them, or build them visually. Tune the limits to your setup.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
ENFORCE POLICY ON PROAGENTSTORE →

Instant setup, no code required.

FULL CATALOGUE

All 243 ProAgentStore tools

FINANCIAL 2 tools
Financial cancel_instance Cancel your subscription and deactivate one private subscribed instance. Financial set_budget_limits Patch your per-account AI-spend limits — the two daily circuit breakers and the four per-tree run knobs. Only
DESTRUCTIVE 22 tools
Destructive cancel_instance_task Cancel a task on the registered local or managed browser runtime for a private instance. Destructive clear_finished_tasks clear_finished_tasks Destructive clear_instance_messages Clear a subscribed instance Destructive clear_instance_voice_settings Drop a subscribed instance Destructive coding_loop_queue_cancel Withdraw a queued objective before it starts. Takes the Destructive coding_repo_remove Detach a repo from a coding instance — the counterpart to coding_repo_add, and the way to clean up a binding t Destructive delete_agent Permanently delete one of YOUR OWN agent TEMPLATES. This also removes the creator Destructive delete_agent_knowledge Permanently remove one knowledge document from an agent TEMPLATE. This changes what new subscribers can receiv Destructive delete_agent_task Remove a standing task from an instance Destructive delete_connection Delete an event connection for good. This destroys the edge Destructive delete_feedback Permanently delete one feedback row — the owner Destructive delete_instance_connector_grant Revoke an instance Destructive delete_instance_file Delete an uploaded file from a subscribed instance (Knowledge → Files). Removes the R2 object, its metadata, a Destructive delete_instance_knowledge Delete a knowledge document from your private subscribed instance. Destructive delete_instance_memory Delete one memory entry (by key) from a subscribed instance. Destructive delete_instance_message Permanently delete the conversational turn containing one message from a subscribed instance. This can remove Destructive delete_instance_task Delete a ticket from a private instance Destructive delete_instance_trigger Delete a configured instance trigger. This removes its event history and connector sync ledger. Destructive delete_supervision Remove a supervision link so the supervisor can no longer delegate to that agent. Destructive and silent: afte Destructive forget_runner_node Permanently remove a disconnected machine Destructive remove_repo Remove one indexed repository from a repo-chat instance (by repo_url or owner/repo), or all of them if neither Destructive unregister_instance_runtime Remove the registered runtime endpoint for one of your private instances.
EXECUTE 32 tools
Execute answer_instance_input Answer a needs_input handoff: an agent paused mid-run because it needs a value it must not invent (work author Execute answer_instance_mcp_input_request Answer or cancel a paused outbound MCP call. Execute apply_to_job Launch the LLM-driven job application for a private apply-agent instance: the PAGS agent drives the user Execute approve_instance_task Approve a browser runtime task waiting for human approval. Cannot approve a connector call held back by the pe Execute call_instance_tool Invoke a connector tool (e.g. github_workflow_runs, github_list_issues) on one of your instances. This is a tw Execute chat_with_agent Send a message to a published agent (trial mode) Execute chat_with_instance Chat with your private subscribed instance of an agent. This is the real runtime path with user-owned state an Execute coding_engine_reauth Sign a coding engine back in from any device when it is blocked on its own login ( Execute coding_loop_start Give an agent an objective and let it work on it autonomously, on the server. Returns a run id immediately — p Execute coding_loop_stop Ask an autonomous run on an instance to stop. Cooperative: the step in flight finishes and settles its spend. Execute coding_overseer Ask the cross-repo Overseer agent. It sees all repos, their live sessions, and recent terminal output. Can ans Execute coding_session_end End a coding session completely. Stops the CLI process/session on the runner. Execute coding_session_fresh End the current session and start a brand new one (clean state, no --resume). Fixes corrupted CLI state. Execute coding_session_open Pick up work on a repo: opens its conversation and CONTINUES where the last one left off (the engine keeps the Execute coding_session_restart Restart the coding CLI session on the selected runner node. Use when the CLI is stuck or erroring. Execute continue_instance_run Carry a STOPPED run Execute end_instance_takeover End a human-takeover session, handing the browser back. This does NOT resume the run — use resume_instance_tak Execute execute_agent_builder_plan Execute a reviewed agent-builder plan to create or scaffold a new agent template. This can create durable plat Execute force_runner_attach Force a machine Execute pause_instance Temporarily stop one private instance WITHOUT unsubscribing: blocks new runs from starting, and asks any run i Execute replay_connection_delivery Re-arm ONE dead delivery so the pump tries it again — the escape hatch for \ Execute resume_instance Put a PAUSED instance back to work: new runs may start again, and background work (cron, the pump, delivery re Execute resume_instance_takeover Resume a run after a human dealt with what it handed off — a captcha solved, a widget operated, a page signed Execute run_instance_task Create a task on the registered local or managed browser runtime for a private instance. The PAGS brain stays Execute run_instance_trigger Manually run one configured instance trigger now. This can create tasks, add knowledge, or sync a granted conn Execute send_instance_takeover_input Send ONE mouse or keyboard event into a taken-over page (relayed to the real browser over CDP). This is the ra Execute set_instance_runner_node Pin one instance to a specific machine, so its runner calls (chat tools, apply, coding) route there — and MOVE Execute start_instance_browser_task Start a browser task on a private instance: the agent drives the owner Execute start_instance_loop Give an agent an objective and let it work autonomously on the server. Durable: it survives you closing the br Execute stop_instance_loop Ask an autonomous run to stop. Cooperative: the step in flight finishes and settles its spend rather than bein Execute test_instance_mcp_server Connect to a remote MCP endpoint as this instance would and report what is actually true about it: whether it Execute trigger_agent_deploy Trigger the GitHub Actions deploy workflow for an owned agent repo.
WRITE 67 tools
Write add_instance_knowledge Add user-specific knowledge to your private subscribed instance. This does not alter the creator Write add_knowledge Add a document to an agent Write apply_account_coding_default Apply your saved account-wide coding CLI default to reachable idle coder sessions now. This is the MCP equival Write batch_write_agent_files Create or overwrite multiple files in an owned agent Write chat_with_my_agent Send a test message to one of YOUR OWN agent TEMPLATES. This is the creator conversation returned by agent_mes Write coding_engine_set Choose which coding CLI this instance opens sessions with, and optionally the model it runs — e.g. switch to a Write coding_repo_add Add a repo to a coding instance. A coding repo is BOTH a local checkout the engine runs in AND its GitHub iden Write create_agent Create a new agent on ProAgentStore. Write create_agent_task Add a standing task to an instance Write create_agent_version Save the current state of one of YOUR OWN agent TEMPLATES as an immutable version snapshot. This does not chan Write create_collection Create a new data collection for an agent. Define fields with types and indexing. Write create_connection Route an event one agent emits to another agent — the Write create_instance_ticket Put a ticket on an instance Write create_instance_trigger Create a webhook, cron, or connector-sync trigger on a subscribed private instance. Use sync_connector with co Write create_supervision Put one agent in charge of another: the supervisor may then delegate goals to the subordinate. Rejected if the Write import_instance_drive_file Copy one Google Drive file into an instance Write import_instance_workdrive_file Copy one Zoho WorkDrive file into an instance Write ingest_repo Index a GitHub repository into a read-only repo-chat instance (the Write insert_instance_record Insert a new record into a collection on one of your subscribed instances. Respects the collection Write insert_record Insert a new record into a collection. Write mark_all_notifications_read Mark EVERY unread notification in your feed read — account-wide, across all instances, not just one page of li Write mark_notification_read Mark ONE notification read, by the Write plan_agent_builder Turn a natural-language request into the deterministic agent-builder plan the creator UI uses: proposed templa Write promote_board_item Make a board card a first-class ticket (757): its runs are then stored as the ticket Write register_instance_runtime Register a local or managed ProAgentStore browser runtime for one of your private instances. Use this before r Write rename_instance Set (or clear) a subscribed instance Write resolve_feedback Close the loop on one piece of feedback: set its status and record the issue it became. This is what keeps the Write rollback_agent_version Replace an agent TEMPLATE Write runner_setup Read the setup checklist for a coding agent that runs on your own machine (e.g. a Write runner_update Update a machine Write scaffold_agent Create a ProAgentStore agent and scaffold its GitHub repo from a starter template. Requires PAGS token plus GI Write set_account_preferences Change your account-wide preferences. Patches by SECTION: pass only the sections you mean to change and the re Write set_agent_capabilities Replace the supplied declared capabilities of an agent TEMPLATE: surfaces, runtime, workflow, tool allowlist a Write set_agent_settings_schema Replace an agent Write set_agent_state Replace the supplied Durable Object state of an agent TEMPLATE — identity, personality, goal, guardrails and w Write set_agent_stats_schema Replace an agent Write set_board_item_status Move a board card to a different column (or reset it to automation by omitting status). Get valid statuses fro Write set_connection_enabled Pause or resume an event connection without deleting it. While paused the source Write set_instance_board_config Customize a private instance Write set_instance_connector_account Pin an instance to ONE of your accounts on a connector, so its calls through that connector stop refusing with Write set_instance_connector_consent Set WRITE consent for one connector on one instance — the owner Write set_instance_instructions Replace a subscribed instance Write set_instance_loop_limits Set the iteration floor and ceiling for an instance Write set_instance_loop_presets Save an instance Write set_instance_mcp_grant Grant or revoke ONE remote MCP tool on ONE endpoint for an instance. Write set_instance_model Pick the model that runs a subscribed instance Write set_instance_operator_manual Replace the operator manual for a subscribed instance (max 16000 chars). The manual is caller-facing guidance Write set_instance_settings Update a subscribed instance Write set_instance_stats Add, edit or remove stats cards on YOUR subscribed instance. Patch semantics: pass ops as [{id, card}]; Write set_instance_terminal_session Save the terminal session the console Write set_instance_tool Switch one tool on or off for one of your instances — the owner Write set_instance_voice_settings Customise a subscribed instance Write set_supervision_enabled Pause or resume a supervision link without deleting it. While paused the supervisor cannot delegate to that ag Write set_translation_config Update a subscribed instance Write subscribe_agent Subscribe to a published agent and create your own private runnable instance. Use this before chat_with_instan Write update_agent Update an agent Write update_agent_board_config Update the authenticated creator Write update_agent_task Edit one standing task — its title, description, or Write update_board_ticket Amend an existing board ticket Write update_instance_knowledge Edit one knowledge document on your private subscribed instance IN PLACE — same Write update_instance_record Change fields on ONE existing record in a collection on one of your subscribed instances. The fields you pass Write update_profile Update your structured candidate Profile / Job Preferences (string fields only; used by the apply pipeline). R Write update_record Update fields on an existing record. Write upload_agent_file Upload a file to an agent Write upload_resume Upload/replace the candidate Write write_agent_file Create or overwrite a file in an owned agent Write write_instance_memory Create or update a memory entry on a subscribed instance. Read get_instance_memory first to reuse an existing
READ 119 tools
Read account_activity What every instance on this account is doing right now, in one call — Read agent_activity Get recent activity log for an agent (chat, tool calls, file uploads, etc.). Read agent_analytics Get usage analytics for an agent Read agent_deploy_status Check the latest GitHub Actions deploy runs for an agent repo. Read agent_info Get an agent Read agent_messages The TEMPLATE Read agent_trace Reconstruct the complete, time-ordered timeline of what an agent instance DID — chat turns (chat.in/tool.call/ Read ask_ticket Ask ONE board ticket a question and get the agent Read billing_status Read your billing/plan status (free vs Pro, whether the paywall is enforced, whether a billing account exists) Read check_instance_loop Check an autonomous run: status, how many steps it has taken, and why it stopped. Omit run_id to list recent r Read coding_diagnostics Full diagnostics for a coding instance: runner connectivity, terminal sessions, repos, issues. Runner state is Read coding_engine_get Which coding CLI this instance opens sessions with, and which model it runs. Answers Read coding_instance_deploy_status Latest GitHub Actions workflow runs for a coding instance Read coding_loop_queue List the objectives queued behind an instance Read coding_loop_status coding_loop_status Read coding_loop_trace Read the live trace for one autonomous coding loop run by Read coding_repos_list List all repos registered in a coding instance, with their status and active sessions. Answers Read coding_session_capture coding_session_capture Read coding_session_message coding_session_message Read coding_sessions_list List all coding sessions (active + ended) for an instance. Read coding_terminal Read a coding session Read coding_timeline Read what a coding run is DOING, while it is still running — the objective it was given, each instruction sent Read connector_status Is a file connector connected for your account, and is this deployment configured for it? Returns Read email_status Gmail connection status for the email-access tool (configured? connected?). Connect/disconnect happens in the Read error_summary What is RECURRING in the platform error log, rather than what happened last — ask this before list_errors when Read export_agent A complete JSON backup of an agent TEMPLATE: its catalogue fields, its DO state, and EVERY knowledge document Read get_account_preferences Read your account-wide preferences — the defaults that follow you across every agent: Read get_agent_board_config Read the authenticated creator Read get_agent_capabilities What an agent TEMPLATE declares it can do: Read get_agent_memory The TEMPLATE Read get_agent_settings_schema Read an agent Read get_agent_state The TEMPLATE Read get_agent_stats_schema Read an agent Read get_apply_tips Read the learned per-ATS apply tips for an apply-capable instance (what worked/failed on each ATS host — conso Read get_budget_limits Read your AI-spend circuit breakers: effective daily ceilings (charged micros + token count), which tier each Read get_creator_dashboard Read your creator dashboard: how many agents you own, their active subscribers, total recorded usage, and the Read get_instance_behaviour_schema Read the platform Read get_instance_board_config Read a private instance Read get_instance_connection_guide The ready-made briefing for driving ONE instance: its id, its agent type, its registered repos, the tools it a Read get_instance_connector_account Which of your accounts an instance uses, per connector you hold an account on (e.g. Gmail). Each row carries t Read get_instance_instructions Read a subscribed instance Read get_instance_loop_limits Read the iteration floor and ceiling configured on an instance — the bounds the platform applies to EVERY run Read get_instance_loop_presets Read the saved objectives an instance offers when a loop starts — each Read get_instance_memory Read a subscribed instance Read get_instance_operator_manual Read the operator manual for a subscribed instance — caller-facing notes about HOW this instance is meant to b Read get_instance_pipeline Read back the stored definition of a named declarative pipeline on this instance (issue 464). Returns the full Read get_instance_settings Read a subscribed instance Read get_instance_state Read a subscribed instance Read get_instance_stats Read a subscribed instance Read get_instance_task Read ONE task/ticket on a private instance in full — status, title, description, and whatever the runtime hold Read get_instance_terminal_session Read the terminal session last selected for one instance. Read get_instance_voice_settings Read a subscribed instance Read get_profile Read the authenticated user Read get_translation_config Read a subscribed instance Read get_usage_dashboard Read your subscriber usage dashboard: active instances, chat and AI-call totals, plus the last 30 days by day Read grant_instance_connector_folder Grant one of your instances access to a folder on a file connector, by share URL or resource id. Returns the c Read hint_instance_task Attach a hint to a runtime task (guidance the agent reads on its next step, e.g. answering a blocked task Read ingest_instance_knowledge_url Fetch a public web page into your private subscribed instance Read ingest_repo_status List the repositories indexed on a repo-chat instance and each one Read instance_activity Read a subscribed instance Read instance_board Read a private instance Read instance_messages Read recent messages from one of your private subscribed instances, newest page first. The response carries Read instance_runner_node Read which machine ONE instance is pinned to, and which machines it could be pinned to. Read instance_runtime_status Check the registered local or managed browser runtime for one of your private instances. Read instance_task_events Read recent events from a private instance Read keys_status Which AI providers have a BYOK key stored for your account (names only — values are never exposed). Useful whe Read list_agent_files List files stored by an agent (resumes, documents, etc.). Read list_agent_repo_files List files in an owned agent Read list_agent_tasks The standing tasks an instance carries in its own task store — the ones injected into its system prompt every Read list_agents List all published agents on ProAgentStore Read list_collections List all data collections (tables) for an agent. Shows schema and record counts. Read list_connection_deliveries The pump Read list_connections List the agent-to-agent event connections leaving an instance — how a fact it emits (e.g. lead.created) is rou Read list_connectors List every connector this deployment knows, resolved for the signed-in account: id, label, auth kind, scopes, Read list_errors Read the platform error log — persisted failures (key-proxy, sign-in, apply/coding, and workflow crashes) that Read list_feedback Read what the OWNER said went wrong (514) — in-session complaints about an agent, each anchored to the turn it Read list_instance_collections List all data collections (tables) for one of your subscribed instances. Shows schema and record counts. Read list_instance_connector_consents Read the owner-granted WRITE consent rows for one instance. Each row names its Read list_instance_connector_grants List the folders one of your instances has been granted on a file connector. Each grant carries the Read list_instance_connectors What THIS instance may do with each connector — the per-agent verdict, not the account-level connection. An ac Read list_instance_drive_files List the files inside a Google Drive folder one of your instances has been granted. Read list_instance_files List files uploaded to a private subscribed instance (PDFs, documents — the console Read list_instance_knowledge List user-specific knowledge documents in your private subscribed instance. Read list_instance_mcp_grants The outbound-MCP grants on one instance: which remote tools, on which endpoints, this agent is allowed to CALL Read list_instance_mcp_input_requests Outbound MCP calls that PAUSED because the remote server asked the person for more (an Read list_instance_tools Audit exactly what one of your instances may do. Returns EVERY tool it could run — its built-in agent faciliti Read list_instance_trigger_events Read recent event history for one configured instance trigger, including received/running/succeeded/failed ent Read list_instance_triggers List webhook, cron, and connector-sync triggers configured on a subscribed private instance. Read list_knowledge List documents in an agent Read list_mcp_presets First-party MCP servers this deployment knows about, as prefilled URLs. A preset is a URL and nothing more: it Read list_notifications Read your notification feed — what your agents and runs have been telling you (a run finished, stopped, or nee Read list_pipeline_runs List an instance Read list_runner_nodes List every machine of yours running a ProAgentStore CLI ( Read list_stats_sources Read the supported stats-card source catalogue: each source Read list_supervision List the agents a supervisor instance oversees (its direct reports). Supervision is how one agent delegates go Read list_trigger_actions The trigger action vocabulary, annotated for ONE instance: every action with Read mcp_audit_log Read recent MCP write, runtime, dry-run, denied, and destructive tool audit events for the authenticated accou Read my_agent Read one of YOUR OWN agent templates in full, as its owner — including a DRAFT, which agent_info cannot see at Read my_agents List agents owned by the authenticated ProAgentStore creator. Read my_instances List your subscribed runnable agent instances. These are the correct targets for real agent chats. Paused inst Read platform_guide Get ProAgentStore platform guide Read platform_health Is the platform healthy right now, and where is the time going? Read-only, safe to call at any time, no side e Read preview_instance_run_continue What continuing a STOPPED run would actually carry forward, before spending anything on it (806). Answers with Read preview_instance_trigger Check a trigger BEFORE creating it. Returns Read query_instance_records Query records from a collection on one of your subscribed instances. Filter by field values, sort, paginate. Read query_records Query records from a collection. Filter by field values, sort, paginate. Read read_agent_file Read a file from an owned agent Read recent_instances The instances THIS account is driving or drove most recently over MCP, each with its latest run Read record_instance_feedback Record something the OWNER says one of their agents got wrong (or did well) — the same row the console Read resync_instance_personality Replace a subscribed instance Read runner_node_forget_preflight Safely inspect whether a machine registration can be forgotten. Returns every alias that would be removed, whe Read sdk_reference Get ProAgentStore SDK usage examples Read search_agent_knowledge Semantic search across an agent Read search_instance_knowledge Semantic (vector) search across a private instance Read system_status Full diagnostics for a coding instance: runner connectivity, node name, tmux sessions, repos, issues — and Read ticket_thread Read the question-and-answer thread on ONE board ticket, oldest first (150). Use it before ask_ticket to see w Read usage_summary Token usage + ESTIMATED value across all your agents over a time Read vector_stats What Read whoami Which account this connection is signed in as: id, login, sign-in provider — the string
OTHER 1 tools
Other name name
EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

UnClick 1662 tools
adminautomation
Apibase 1384 tools
adminautomation
Delx Mcp A2a 1068 tools
adminautomation
MCP Framework Personal 912 tools
adminautomation
Hermoso 896 tools
adminautomation
Nodebench 824 tools
adminautomation
Altium-MCP-Suite 817 tools
adminautomation
Eda Agent 777 tools
adminautomation
FAQ

Questions about ProAgentStore

Can an AI agent move money through the ProAgentStore MCP server? +

Yes. The ProAgentStore server exposes 2 financial tools including cancel_instance, set_budget_limits. Without a policy, an autonomous agent can call these with no spend caps, no rate limits, and no approval flow. PolicyLayer lets you block financial tools by default, require human approval, or set per-tool rate limits — enforced on every call.

Can an AI agent delete data through the ProAgentStore MCP server? +

Yes. The ProAgentStore server exposes 22 destructive tools including cancel_instance_task, clear_finished_tasks, clear_instance_messages. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through ProAgentStore? +

The ProAgentStore server has 67 write tools including add_instance_knowledge, add_knowledge, apply_account_coding_default. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach ProAgentStore.

How many tools does the ProAgentStore MCP server expose? +

243 tools across 4 categories: Destructive, Execute, Read, Write. 119 are read-only. 124 can modify, create, or delete data.

How do I enforce a policy on ProAgentStore? +

Register the ProAgentStore MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every ProAgentStore tool call.

Deterministic rules across all 243 ProAgentStore tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON PROAGENTSTORE →

Instant setup, no code required.

243 ProAgentStore tools catalogued and risk-classified — across an index of 46,500+ MCP servers.

// WHERE THIS COMES FROM

These policies come from ProAgentStore's registry record.

The record behind this page: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.

Teams ship this data inside their own products. See what a licence covers →

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.