Feature requests

Discover our plans and suggest new improvements to Speakeasy.

Agent Session Portability

Remote Authentication Branding

Enterprise managed authorization for MCP

MCP Risk Approval

OAuth CIMD Support

Platform MCP and Plugin

Streaming and Bulk export of data feeds

[Risk UX] Watchdog: Signals, Ranking, Triage

End User Tool Filtering

Prompt Based Policies: Policy Suggest, Hybrid & Session Quarantine

Multiple MCP endpoint support

Cost: Subscriptions

Shadow AI Monitoring

Antigravity coverage

Zed coverage

Security: Secret Backends

Internal MCP Server Portals

MCP staging and versioning

Gemini Support

URL Mode elicitation for Remote Sessions

Security: Tool Call Redaction

Cost: Startup Tokens Capture

Cost: Auto-Categorisation of Work Type

Assistants Beta Release

Device Agent: Request access for blocked AI tool calls

Prompt Injections Scanning in Skills

Prompt Based Policies: Policy Evals (Session Replay)

Platform-level remote identity providers

Tool Drift Detection

Multiple Remote Sessions per MCP Server

BYOK for Managed Inference (OpenRouter)

OpenCode support

Security: Confirm / Warn policy type

MCP Network-Level Access Control

Gram: Support Custom Code for Tools (aka "Gram Functions)

Gram: Support spec uploads without operationIds

Gram: CLI v0.2.0

Gram: Faster Deployments

Gram: Customer Configured MCP Install Page

Remote MCP Support

Skills Management

MCP Logs dashboard improvements

Shadow MCP Request Approval Flow

Security: Prompt based policies

Environment UX Improvements

Workforce Observability Dashboard

RBAC for risk policies

MCP Server Tool Filtering

Support organization-level remote session issuers

Requests & Approvals

Anthropic Compliance API Integration

Exclusions and retroactive suppression of risk findings

Personal Account Tracking

Private MCP Tunnels

Cost: Tool- & MCP-level Cost Tracking

Shadow MCP Inventory

Risk Finding Severity

Shadow MCP Policy Gates

LiteLLM Usage Integration

Feature requests

Discover our plans and suggest new improvements to Speakeasy.

Agent Session Portability

Remote Authentication Branding

Enterprise managed authorization for MCP

MCP Risk Approval

OAuth CIMD Support

Platform MCP and Plugin

Streaming and Bulk export of data feeds

[Risk UX] Watchdog: Signals, Ranking, Triage

End User Tool Filtering

Prompt Based Policies: Policy Suggest, Hybrid & Session Quarantine

Multiple MCP endpoint support

Cost: Subscriptions

Shadow AI Monitoring

Antigravity coverage

Zed coverage

Security: Secret Backends

Internal MCP Server Portals

MCP staging and versioning

Gemini Support

URL Mode elicitation for Remote Sessions

Security: Tool Call Redaction

Cost: Startup Tokens Capture

Cost: Auto-Categorisation of Work Type

Assistants Beta Release

Device Agent: Request access for blocked AI tool calls

Prompt Injections Scanning in Skills

Prompt Based Policies: Policy Evals (Session Replay)

Platform-level remote identity providers

Tool Drift Detection

Multiple Remote Sessions per MCP Server

BYOK for Managed Inference (OpenRouter)

OpenCode support

Security: Confirm / Warn policy type

MCP Network-Level Access Control

Gram: Support Custom Code for Tools (aka "Gram Functions)

Gram: Support spec uploads without operationIds

Gram: CLI v0.2.0

Gram: Faster Deployments

Gram: Customer Configured MCP Install Page

Remote MCP Support

Skills Management

MCP Logs dashboard improvements

Shadow MCP Request Approval Flow

Security: Prompt based policies

Environment UX Improvements

Workforce Observability Dashboard

RBAC for risk policies

MCP Server Tool Filtering

Support organization-level remote session issuers

Requests & Approvals

Anthropic Compliance API Integration

Exclusions and retroactive suppression of risk findings

Personal Account Tracking

Private MCP Tunnels

Cost: Tool- & MCP-level Cost Tracking

Shadow MCP Inventory

Risk Finding Severity

Shadow MCP Policy Gates

LiteLLM Usage Integration

Pre-emptive Remote Session Refresh