MCP 服务器

agentguard — policy proxy for agent tool calls

io.github.agentwares/agentguard-cloud
MCP 与智能体基础设施 安全 公开且可连接 MCP 2025-11-25

此 MCP 可以做什么

Creates policy-controlled proxies for MCP servers and provides spending reports, audit exports, allowlist recommendations, and tool-call policy management.

agentguard_create_proxy
Create a hosted agentguard proxy in front of an MCP server. Returns the proxy id, the per-customer MCP URL and the proxy key (shown once). Point your agent's MCP client at the URL with `Authorization: Bearer <key>`. Requires an account key.
输入模式
{'type': 'object', '$schema': 'https://json-schema.org/draft/2020-12/schema', 'required': ['slug', 'upstreamUrl'], 'properties': {'mode': {'enum': ['dry-run', 'enforce'], 'type': 'string', 'default': 'dry-run', 'description': 'dry-run records what the policy would have blocked and lets every call through; enforce actually blocks. Start in dry-run.'}, 'slug': {'type': 'string', 'maxLength': 40, 'minLength': 2, 'description': 'short name, a-z 0-9 -'}, 'upstreamUrl': {'type': 'string', 'format': 'uri', 'description': 'Streamable HTTP URL of the MCP server to protect'}, 'upstreamName': {'type': 'string', 'maxLength': 32, 'minLength': 1, 'description': 'name used to namespace tools when you add more upstreams later'}}, 'additionalProperties': False}
agentguard_export_audit
Export a run's hash-chained audit entries (prev_hash, hash, redacted args) with the Merkle root, for disputes and questionnaires. Verify offline with sha256(prev_hash + canonical(entry)). Requires an account key.
只读 幂等
输入模式
{'type': 'object', '$schema': 'https://json-schema.org/draft/2020-12/schema', 'required': ['proxyId', 'runId'], 'properties': {'runId': {'type': 'string', 'minLength': 1, 'description': 'the X-Run-Id your agent sent with the calls; one run is one agent session'}, 'proxyId': {'type': 'string', 'minLength': 6, 'description': 'the proxy id returned by agentguard_create_proxy'}}, 'additionalProperties': False}
agentguard_get_pricing
Machine-readable pricing for agentguard hosted: bands (Starter/Pro/Team) with included tool calls per month, the 80% soft alert and 150% hard stop, features, the free trial allowance, and `per_call` — buying tool calls outright at POST /api/v1/calls from a prepaid balance, which is the one purchase here an unattended agent can complete. Same data as /pricing.json.
只读 幂等
输入模式
{'type': 'object', '$schema': 'https://json-schema.org/draft/2020-12/schema', 'properties': {}, 'additionalProperties': False}
agentguard_get_run
The incident-shaped report for one run: counts (calls, reads, writes, faked, blocked, spend), the call timeline with decisions, dry-run mutations (what would have changed), ledger entries and the audit-chain verification. Requires an account key.
只读 幂等
输入模式
{'type': 'object', '$schema': 'https://json-schema.org/draft/2020-12/schema', 'required': ['proxyId', 'runId'], 'properties': {'runId': {'type': 'string', 'minLength': 1, 'description': 'the X-Run-Id your agent sent with the calls; one run is one agent session'}, 'proxyId': {'type': 'string', 'minLength': 6, 'description': 'the proxy id returned by agentguard_create_proxy'}}, 'additionalProperties': False}
agentguard_recommend_allowlist
Least-privilege recommendation from the proxy's recorded calls: the minimal tool allowlist, argument keys used per tool, and the upstream tools never used (remove them first). No LLM call here; apply it in the dashboard. Requires an account key.
只读 幂等
输入模式
{'type': 'object', '$schema': 'https://json-schema.org/draft/2020-12/schema', 'required': ['proxyId'], 'properties': {'proxyId': {'type': 'string', 'minLength': 6, 'description': 'the proxy id returned by agentguard_create_proxy'}}, 'additionalProperties': False}
agentguard_set_policy
Replace a proxy's policy (YAML: mode, upstreams, classify, caps, loop, dry_run, approvals, alerts, chaos, drift). Validated before saving; takes effect within 30 seconds. Requires an account key.
幂等
输入模式
{'type': 'object', '$schema': 'https://json-schema.org/draft/2020-12/schema', 'required': ['proxyId', 'policyYaml'], 'properties': {'proxyId': {'type': 'string', 'minLength': 6, 'description': 'the proxy id returned by agentguard_create_proxy'}, 'policyYaml': {'type': 'string', 'maxLength': 64000, 'minLength': 1, 'description': 'the complete policy document; it replaces the current one rather than merging'}}, 'additionalProperties': False}
agentguard_spend_report
Read an agent log and report what it spent: totals, calls by tool and by class, estimated dollars by rail, loops, the largest single run, and — for every overspend — the agentguard band whose cap would have stopped it, the exact cap, and the call that would have tripped it. Needs no account and no key. The log is read in the request and discarded; nothing is stored. Accepts an agentwares audit export (`agentwares.audit/v1`), a Claude Code session `.jsonl`, or a CSV/JSONL of tool calls with any of `tool`, `ts`, `class`, `amount_usd`, `rail`, `run_id`, `model`, `input_tokens`, `output_tokens`. Caps come from the proxy's own default policy, so a band named here refuses exactly what it says it refuses. Buying that band is a checkout a person completes; the calls-only purchase an unattended agent can complete is in `agentguard_get_pricing` instead.
只读 幂等
输入模式
{'type': 'object', '$schema': 'https://json-schema.org/draft/2020-12/schema', 'required': ['log'], 'properties': {'log': {'type': 'string', 'minLength': 1, 'description': 'The log itself, verbatim — JSON, JSONL or CSV. Not a path and not a URL: this server does not fetch anything.'}, 'detail': {'enum': ['concise', 'full'], 'type': 'string', 'description': '`concise` (default) returns the totals, the top ten tools, the rails, the loops and every overspend. `full` adds the complete per-tool breakdown and the window. Ask for `full` only when you are going to read it.'}}, 'additionalProperties': False}
已添加
agentguard_recommend_allowlist
2026年9月17日 12:40
已添加
agentguard_export_audit
2026年9月17日 12:40
已添加
agentguard_get_run
2026年9月17日 12:40
已添加
agentguard_set_policy
2026年9月17日 12:40
已添加
agentguard_create_proxy
2026年9月17日 12:40
已添加
agentguard_spend_report
2026年9月17日 12:40
已添加
agentguard_get_pricing
2026年9月17日 12:40