⚙️ [STEP 3/5: BUILD] Process pending sources into searchable chunks. PREREQUISITES: - project_create must have been run - project_add_source must have added at least one source CHUNKING OPTIONS (optional chunk_options): - max_sources_per_build: Process N sources per call (default: 10) - fetch_con...
AI agents invoke indexfoundry_project_build to trigger actions in IndexFoundry 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 executes a multi-step pipeline: fetching external content, parsing/extracting structured data, and transforming it into vector database entries. While it does not permanently destroy data (not Destructive) or make irreversible changes beyond the build artifact itself, it actively triggers external operations (URL fetches with configurable concurrency, PDF extraction, HTML parsing) whose effects depend on…
From the tool's definition The tool description states it will 'Fetch content from pending sources' and 'Extracts text' and 'Chunks text', which are operations that trigger external data processing and transformation.
Attacks that exploit this kind of access
⚙️ [STEP 3/5: BUILD] Process pending sources into searchable chunks. PREREQUISITES: - project_create must have been run - project_add_source must have added at least one source CHUNKING OPTIONS (optional chunk_options): - max_sources_per_build: Process N sources per call (default: 10) - fetch_concurrency: Parallel URL fetches (default: 3) - enable_checkpointing: Enable resume capability (default: true) WHAT THIS DOES: 1. Fetches content from pending sources (up to max_sources_per_build) 2. Extracts text (HTML parsing, PDF extraction, etc.) 3. Chunks text with overlap for context continuity 4. Generates embeddings using OpenAI API (requires OPENAI_API_KEY) 5. Saves checkpoint after each source for resumability 6. Returns progress with remaining sources count RESUME: Use resume_from_checkpoint=true to continue after timeout/failure COST: ~$0.02 per 1M tokens embedded (text-embedding-3-small) NEXT STEPS: - If has_more=true: call project_build again to continue - Use project_build_status to check checkpoint state - Use project_query to test search quality. It is categorised as a Execute tool in the IndexFoundry MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the IndexFoundry MCP server in PolicyLayer and add a rule for indexfoundry_project_build: 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 IndexFoundry MCP. Nothing to install.
indexfoundry_project_build 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 indexfoundry_project_build 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 indexfoundry_project_build. 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.
indexfoundry_project_build is provided by the IndexFoundry MCP server (mnehmos/mnehmos.index-foundry.mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →