AI agents invoke instruct-developer to trigger actions in Agentic Developer MCP. 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.
The server context strongly implies this tool executes coding instructions via Codex CLI, which can run arbitrary code and shell commands. The sibling tool 'clone_and_write_prompt' further suggests write/execute operations. However, the empty description forces a lower confidence — the tool could be read-only analysis or could perform destructive operations.
From the tool's definition Tool name 'instruct-developer' on a server described as wrapping 'OpenAI's Codex CLI to automate repository cloning and code analysis tasks' that 'execute complex coding requests on specific Git branches and subfolders'
Documented attack patterns abuse exactly the kind of access instruct-developer gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Agentic Developer MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for instruct-developer:
{
"version": "1",
"default": "deny",
"tools": {
"instruct-developer": {
"limits": [
{
"counter": "instruct-developer_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}instruct-developer 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.
Free to start. No card required.
instruct-developer. It is categorised as a Execute tool in the Agentic Developer MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Agentic Developer MCP server in PolicyLayer and add a rule for instruct-developer: 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 Agentic Developer MCP. Nothing to install.
instruct-developer 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 instruct-developer 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 instruct-developer. 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.
instruct-developer is provided by the Agentic Developer MCP server (teabranch/agentic-developer-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 3 Agentic Developer MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
3 Agentic Developer MCP tools catalogued and risk-classified — across an index of 42,500+ MCP servers.