Run the full LLM summarization pass over all un-summarized chunks in the index.
AI agents invoke write_summaries to trigger actions in Code-Index-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.
This tool triggers an active processing operation (an LLM summarization pass) over indexed code chunks. It's not a simple read/query — it executes a computational pipeline that modifies the index by writing summaries to previously un-summarized chunks. The most severe applicable category is Execute (triggers external operations), though it also has Write characteristics.
From the tool's definition "Run the full LLM summarization pass over all un-summarized chunks in the index"
Documented attack patterns abuse exactly the kind of access write_summaries gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Code-Index-MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for write_summaries:
{
"version": "1",
"default": "deny",
"tools": {
"write_summaries": {
"limits": [
{
"counter": "write_summaries_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}write_summaries 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.
Run the full LLM summarization pass over all un-summarized chunks in the index. It is categorised as a Execute tool in the Code-Index-MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Code-Index- MCP server in PolicyLayer and add a rule for write_summaries: 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 Code-Index-MCP. Nothing to install.
write_summaries 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 write_summaries 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 write_summaries. 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.
write_summaries is provided by the Code-Index- MCP server (viperjuice/code-index-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 8 Code-Index-MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
8 Code-Index-MCP tools catalogued and risk-classified — across an index of 42,500+ MCP servers.