CONTEXT LOADING FALLBACK: If garden_find returns thin results on a session that implies prior work or continuity, follow with garden_forage to search note bodies. The pair covers intent-based context loading: garden_find first (title match), garden_forage second (body search).\n\nUse this wheneve...
AI agents call garden_forage to retrieve information from Synapse without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
garden_forage retrieves and queries data from an Obsidian vault with no side effects. It searches note bodies and returns matches—a pure read operation. The tool does not create, modify, delete, or execute anything; it only locates and surfaces existing information. This is classic Read category behavior (search, find, query).
From the tool's definition Tool searches note bodies and returns matching files with line numbers and context. Description explicitly states it is a FALLBACK for searching note content when 'garden_find returns thin results'.
Attacks that exploit this kind of access
CONTEXT LOADING FALLBACK: If garden_find returns thin results on a session that implies prior work or continuity, follow with garden_forage to search note bodies. The pair covers intent-based context loading: garden_find first (title match), garden_forage second (body search).\n\nUse this whenever the user wants to search inside their notes for a specific phrase, keyword, quote, or concept. Returns matching files with line numbers and context (case-insensitive). Triggers:. It is categorised as a Read tool in the Synapse MCP Server, which means it retrieves data without modifying state.
Register the Synapse MCP server in PolicyLayer and add a rule for garden_forage: 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 Synapse. Nothing to install.
garden_forage is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the garden_forage 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 garden_forage. 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.
garden_forage is provided by the Synapse MCP server (tomjrworks/synapse-obsidian). 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 →