// SCAN REPORT

OpenSpec MCP

40 tools. 16 can modify or destroy data without limits.

16 write tools that can modify data. Rate limits recommended.

Last updated: 11 Jun 2026

16 can modify or destroy data

24 read-only

40 tools total

Community server · catalogue entry verified 11/06/2026

How to control OpenSpec MCP ↓

TOOL MAP

What OpenSpec MCP exposes to your agents

Read (24) Write / Execute (16) Destructive / Financial (0)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

High Risk

The most dangerous OpenSpec MCP tools

Execute · High openspec_setup_hooks Installing or uninstalling Git hooks modifies the repository's hook scripts, which are executable scripts that run automatically on Git events. Write · Medium openspec_batch_update_tasks This tool modifies task statuses within a change proposal, which is reversible state manipulation. Write · Medium openspec_add_review This tool creates new review comments, which are reversible modifications to a change or spec entity. Write · Medium openspec_approve_change This tool modifies the state of a change by marking it approved, which is a reversible write operation (approval can typically be undone or superseded by new review).

16 of OpenSpec MCP's 40 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control OpenSpec MCP

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

Rate limit write operations

{
  "openspec_record_revision": {
    "limits": [
      {
        "counter": "openspec_record_revision_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "openspec_ai_analyze_context": {
    "limits": [
      {
        "counter": "openspec_ai_analyze_context_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

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

Create a free account and register OpenSpec MCP — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON OPENSPEC →

Free to start. No card required.

FULL CATALOGUE

All 40 OpenSpec MCP tools

EXECUTE 1 tools

Execute openspec_setup_hooks Install or uninstall OpenSpec Git hooks

WRITE 15 tools

Write openspec_record_revision Record a design revision when fixing bugs or adjusting design that deviates from the original proposal/design. Write openspec_reply_review Reply to a review comment Write openspec_request_approval Request approval for a change Write openspec_sync_revisions Sync recorded revisions to design.md or proposal.md before archiving Write openspec_add_review Add a review comment to a change or spec Write openspec_approve_change Approve a change Write openspec_archive_change Archive a completed change proposal. Automatically syncs revisions to design.md before archiving. Write openspec_batch_update_tasks Batch update multiple task statuses in a change Write openspec_create_change Create a new change from a template Write openspec_generate_proposal Generate a proposal from requirement description (returns template for AI to complete) Write openspec_reject_change Reject a change Write openspec_resolve_review Mark a review as resolved or won\ Write openspec_save_proposal Save a generated proposal and tasks to a new change Write openspec_update_revision Update an existing revision, usually to add metadata or correct information. Write openspec_update_task Update task status

READ 24 tools

Read openspec_ai_analyze_context 使用 AI 深度分析项目上下文并给出 project.md 更新建议。利用 MCP Client 的 AI 能力进行分析。 Read openspec_analyze_context 分析项目上下文（技术栈、目录结构、代码模式） Read openspec_critique_proposal 评审 proposal 或 design 文档，识别潜在问题（完整性、可行性、安全、边界条件、清晰度） Read openspec_get_approval_status Get approval status for a change Read openspec_get_critique_result 获取变更的评审结果（最新或历史记录） Read openspec_get_instructions Get OpenSpec usage instructions from AGENTS.md Read openspec_get_progress_summary Get progress summary for all changes Read openspec_get_project_context Get project context from project.md Read openspec_get_review_summary Get review summary for a change (all files) Read openspec_get_tasks Get tasks and progress for a change Read openspec_list_changes List all OpenSpec change proposals Read openspec_list_cross_service_docs List cross-service documents for a change proposal. Returns documents defined in proposal.md frontmatter cross Read openspec_list_pending_approvals List all changes pending approval Read openspec_list_reviews List review comments for a change or spec Read openspec_list_revisions List all recorded design revisions for a change Read openspec_list_specs List all OpenSpec specifications Read openspec_list_templates List all available change templates Read openspec_preview_template Preview the content of a template Read openspec_read_cross_service_doc Read a specific cross-service document content. Use openspec_list_cross_service_docs first to see available do Read openspec_show_change Show details of a specific change proposal Read openspec_show_spec Show details of a specific specification Read openspec_validate_all Validate all changes and/or specs Read openspec_validate_change Validate a change proposal Read openspec_validate_spec Validate a specification

FAQ

Questions about OpenSpec MCP

How do I prevent bulk modifications through OpenSpec MCP? +

The OpenSpec MCP server has 15 write tools including openspec_record_revision, openspec_reply_review, openspec_request_approval. 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 OpenSpec MCP.

How many tools does the OpenSpec MCP server expose? +

40 tools across 2 categories: Read, Write. 24 are read-only. 16 can modify, create, or delete data.

How do I enforce a policy on OpenSpec MCP? +

Register the OpenSpec 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 OpenSpec MCP tool call.

Deterministic rules across all 40 OpenSpec MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON OPENSPEC →

Free to start. No card required.

40 OpenSpec MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.