[Feature request] AgentCoreToolSearchPlugin: add tool_filters support to restrict tools resolved from AgentCore Gateway

Open
#567 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
3/5
Estimated time
1-2 days
Newbie friendliness
68/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
python
Domain
ai

Research direction

Start with AgentCoreToolSearchPlugin and its x_amz_bedrock_agentcore_search tool call, then compare how Strands MCPClient handles tool_filters. Add support for an allowed-tool filter so semantic search considers only permitted tools, and verify that AgentCoreToolSearchPlugin accepts the example tool_filters parameter and excludes disallowed results.

Written by the indexing model from the issue text.

Description

Is your feature request related to a problem? Please describe.
When using AgentCoreToolSearchPlugin with AgentCore Gateway, there is currently no way to restrict which tools are surfaced to the agent. The plugin performs semantic search over the full set of tools registered in the Gateway, which can include dozens of tools. Without filtering, the semantic search may surface unintended tools, which increases noise, wastes tokens, and can cause the agent to invoke tools it should not have access to.

Describe the solution you'd like
AgentCoreToolSearchPlugin should support a tool_filters parameter, mirroring the tool_filters parameter already available on Strands' MCPClient:

# Current Strands MCPClient — already supports filtering
MCPClient(
    ...,
    tool_filters={"allowed": ["tool-1", "tool-2"]}
)

# Desired — AgentCoreToolSearchPlugin parity
AgentCoreToolSearchPlugin(
    ...,
    tool_filters={"allowed": ["tool-1", "tool-2"]}
)

Ideally, the filter should be applied before (or at query time during) the semantic search, so that only candidate tools matching the allowlist are considered. If not possible, it can be done after query result I guess inside the x_amz_bedrock_agentcore_search tool call.

Describe alternatives you've considered
Post-filtering after semantic search with the post tool call hook and discarding results that don't match the allowlist. This works but should be natively supported by the plugin.

Additional context
N/A

Dominant language
Python
Stars
764
Forks
149
Avg merge
1d 19h
Merged PRs (30d)
7

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from aws/bedrock-agentcore-sdk-python

All issues in aws/bedrock-agentcore-sdk-python

Similar issues

More Python issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.