http_fetch
ADR-164 §5.1.8 — HTTP probe primitive for business-pod ops benches (synthetic 200/500 endpoint checks, third-party status pages). Default-secure: blocks file://, ftp://, RFC-1918 / loopback / link-local hosts unless CLAUDE_FLOW_HTTP_FETCH_ALLOW_PRIVATE=1, and rejects Authorization / Cookie / X-Au...
This record as markdown: /tools/io-github-ruvnet-claude-flow/http-fetch.md
What http_fetch does on Claude Flow
AI agents invoke http_fetch to trigger actions in Claude Flow. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.
Why http_fetch is rated High
This tool executes outbound HTTP requests to arbitrary URLs, which constitutes triggering external operations. While it has default-secure guards (blocks private IPs, sensitive headers), it can still reach any public endpoint on the internet.
From the tool's definition HTTP probe primitive... synthetic 200/500 endpoint checks, third-party status pages... blocks file://, ftp://, RFC-1918 / loopback / link-local hosts unless CLAUDE_FLOW_HTTP_FETCH_ALLOW_PRIVATE=1...
Attacks that exploit this kind of access
The rule that runs http_fetch safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Claude Flow, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For http_fetch, this is the rule to start with:
http_fetch stays usable, but rate-capped: a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Claude Flow, apply this rule, and every http_fetch call is checked against it from then on.
Questions about http_fetch
ADR-164 §5.1.8 — HTTP probe primitive for business-pod ops benches (synthetic 200/500 endpoint checks, third-party status pages). Default-secure: blocks file://, ftp://, RFC-1918 / loopback / link-local hosts unless CLAUDE_FLOW_HTTP_FETCH_ALLOW_PRIVATE=1, and rejects Authorization / Cookie / X-Auth-* headers unless CLAUDE_FLOW_HTTP_FETCH_ALLOW_AUTH=1. Hard 30s timeout (60s ceiling), response truncated to 256 KB (1 MB ceiling), default User-Agent ruflo-http-fetch/1.0. Use when a pod or smoke contract needs a guarded HTTP probe — calling Node fetch() directly is wrong because it skips the URL allowlist and header sanitization that ADR-164 mandates for autopilot mode. Pair with the ops-pod bench in plugins/ruflo-business-pods/templates/ops.json (the §4.4 synthetic-endpoint test). It is categorised as a Execute tool in the Claude Flow MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Claude Flow MCP server in PolicyLayer and add a rule for http_fetch: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Claude Flow. Nothing to install.
http_fetch is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the http_fetch rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for http_fetch. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
http_fetch is provided by the Claude Flow MCP server (claude-flow). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Claude Flow, and thousands of servers like it.
Across the catalogue