Run a specific named Alexandria agent directly on a task. Use this when you know exactly which specialist you want (e.g., "brand-copywriter", "senior-developer"). Supports two modes (auto-selected via ALEXANDRIA_MODE): (1) DELEGATE — returns the agent's assembled prompt + context + learnings for ...
Accepts file system path (files[].path); Accepts raw HTML/template content (files[].content); Single-target operation
Part of the Alexandria MCP server. Enforce policies on this tool with Intercept, the open-source MCP proxy.
AI agents invoke alexandria_run_agent to trigger processes or run actions in Alexandria. Execute operations can have side effects beyond the immediate call -- triggering builds, sending notifications, or starting workflows. Rate limits and argument validation are essential to prevent runaway execution.
alexandria_run_agent can trigger processes with real-world consequences. An uncontrolled agent might start dozens of builds, send mass notifications, or kick off expensive compute jobs. Intercept enforces rate limits and validates arguments to keep execution within safe bounds.
Execute tools trigger processes. Rate-limit and validate arguments to prevent unintended side effects.
tools:
alexandria_run_agent:
rules:
- action: allow
rate_limit:
max: 10
window: 60
validate:
required_args: trueSee the full Alexandria policy for all 20 tools.
Agents calling execute-class tools like alexandria_run_agent have been implicated in these attack patterns. Read the full case and prevention policy for each:
Other tools in the Execute risk category across the catalogue. The same policy patterns (rate-limit, validate) apply to each.
alexandria_run_agent is one of the high-risk operations in Alexandria. For the full severity-focused view — only the high-risk tools with their recommended policies — see the breakdown for this server, or browse all high-risk tools across every MCP server.
Run a specific named Alexandria agent directly on a task. Use this when you know exactly which specialist you want (e.g., "brand-copywriter", "senior-developer"). Supports two modes (auto-selected via ALEXANDRIA_MODE): (1) DELEGATE — returns the agent's assembled prompt + context + learnings for your next reply to execute. Zero cost, uses your Claude Code subscription. (2) BYOT EXECUTE — runs the agent against your provider key and returns the finished output. Supports optional file context injection for code review, modification, or analysis tasks. For automatic agent selection, use alexandria_ask instead.. It is categorised as a Execute tool in the Alexandria MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Add a rule in your Intercept YAML policy under the tools section for alexandria_run_agent. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the Alexandria MCP server.
alexandria_run_agent 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 alexandria_run_agent rule in your Intercept 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 Intercept policy for alexandria_run_agent. 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.
alexandria_run_agent is provided by the Alexandria MCP server (mcp-server-alexandria). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Open source. One binary. Zero dependencies.
npx -y @policylayer/intercept