Index the current workspace for semantic search. This tool scans all source files in the workspace and builds a semantic index that enables fast, meaning-based code search. When to use this tool: - First time using the context engine with a new project - After making significant changes to the co...
AI agents invoke index_workspace to trigger actions in Context Engine MCP Server. 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.
index_workspace initiates a potentially long-running process that scans files and constructs an index. This is not a simple read (it writes/builds an index) nor purely destructive, but it executes an automated operation with side effects (building/updating the semantic index). It falls under Execute as it triggers an external operation whose scope depends on the workspace contents.
From the tool's definition 'scans all source files in the workspace and builds a semantic index' — triggers an active scanning and indexing operation across the filesystem
Risk signalsAdmin/system-level operation
Documented attack patterns abuse exactly the kind of access index_workspace gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Context Engine MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for index_workspace:
{
"version": "1",
"default": "deny",
"tools": {
"index_workspace": {
"limits": [
{
"counter": "index_workspace_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}index_workspace 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.
Index the current workspace for semantic search. This tool scans all source files in the workspace and builds a semantic index that enables fast, meaning-based code search. When to use this tool: - First time using the context engine with a new project - After making significant changes to the codebase - When semantic_search or enhance_prompt returns no results What gets indexed (50+ file types): - TypeScript/JavaScript (.ts, .tsx, .js, .jsx, .mjs, .cjs) - Python (.py, .pyi) - Flutter/Dart (.dart, .arb) - Go (.go) - Rust (.rs) - Java/Kotlin/Scala (.java, .kt, .kts, .scala) - C/C++ (.c, .cpp, .h, .hpp) - .NET (.cs, .fs) - Swift/Objective-C (.swift, .m) - Web (.vue, .svelte, .astro, .html, .css, .scss) - Config (.json, .yaml, .yml, .toml, .xml, .plist, .gradle) - API schemas (.graphql, .proto) - Shell scripts (.sh, .bash, .ps1) - DevOps (Dockerfile, .tf, Makefile, Jenkinsfile) - Documentation (.md, .txt) What is excluded (optimized for AI context): - Generated code (*.g.dart, *.freezed.dart, *.pb.*) - Dependencies (node_modules, vendor, Pods, .pub-cache) - Build outputs (dist, build, .dart_tool, .next) - Lock files (package-lock.json, pubspec.lock, yarn.lock) - Binary files (images, fonts, media, archives) - Files over 1MB (typically generated or data files) - Secrets (.env, *.key, *.pem) The index is saved to .context-engine-context-state.json in the workspace root and will be automatically restored on future server starts. It is categorised as a Execute tool in the Context Engine MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Context Engine MCP Server MCP server in PolicyLayer and add a rule for index_workspace: 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 Context Engine MCP Server. Nothing to install.
index_workspace 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 index_workspace 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 index_workspace. 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.
index_workspace is provided by the Context Engine MCP Server MCP server (kirachon/context-engine). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Context Engine MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
50 Context Engine MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.