Precompute and cache symbol embeddings for semantic / hybrid search. Embeddings are also computed lazily on first semantic query, but calling this once after a fresh index avoids the first-query latency spike. Requires AI provider to be enabled in config (ollama/openai). Set force=true to drop an...
AI agents use embed_repo to create or update resources in Trace — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Trace environment.
This tool writes/mutates the vector store by computing and caching embeddings. With force=true it drops and recomputes existing data, but this is described as idempotent and the underlying source code is not affected. The primary action is writing to a cache/vector store, not irreversible deletion of user data, so Write is the appropriate category.
From the tool's definition Precompute and cache symbol embeddings... Mutates the vector store; idempotent. Set force=true to drop and recompute all existing embeddings.
Documented attack patterns abuse exactly the kind of access embed_repo gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Trace, and nothing reaches the server without passing your rules. This is the rule we recommend for embed_repo:
{
"version": "1",
"default": "deny",
"tools": {
"embed_repo": {
"limits": [
{
"counter": "embed_repo_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}embed_repo stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Precompute and cache symbol embeddings for semantic / hybrid search. Embeddings are also computed lazily on first semantic query, but calling this once after a fresh index avoids the first-query latency spike. Requires AI provider to be enabled in config (ollama/openai). Set force=true to drop and recompute all existing embeddings. Mutates the vector store; idempotent. Use after reindex when you plan to use semantic search. Returns JSON: { status, indexed_this_run, total_embedded, coverage_pct, duration_ms }. It is categorised as a Write tool in the Trace MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Trace MCP server in PolicyLayer and add a rule for embed_repo: 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 Trace. Nothing to install.
embed_repo is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the embed_repo 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 embed_repo. 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.
embed_repo is provided by the Trace MCP server (nikolai-vysotskyi/trace-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 178 Trace tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
178 Trace tools catalogued and risk-classified — across an index of 42,500+ MCP servers.