AgentIndex icon
AgentIndex
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompareGuides
Submit Tool
ToolsMCP ServersHOTCategoriesTrendingNewBest ListsCompare
Home/
Security & Safety/
phantom-secrets
phantom-secrets logo

phantom-secrets

Active·★ 18·MIT·Updated 2026-10-19

Phantom protects sensitive API keys and credentials when working with AI coding agents by replacing real secrets with scoped placeholders. It uses an authenticated local proxy to inject configured authentication values into API requests, preventing agents from directly accessing or exposing actual secret values.

#AI Agent Security#Secrets Management#Credential Protection#Local Proxy#Vault#API Security
© 2026 AgentIndex.app|Built by a 10-year iOS Developer.
Best by use caseAboutContactAdvertisePrivacy PolicyTerms of UseQYSGitHubBuy me a coffee ☕

Browse by Category

MCP ServersCode AssistantWorkflow AutomationRAG / Knowledge BaseMulti-AgentBrowser AutomationLLM InfraDev ToolingObservability

Not affiliated with Anthropic, OpenAI or Microsoft.

Guides
#Dev Tooling
#Context Isolation
$ Install
$ brew tap ashlrai/phantom && brew trust --formula ashlrai/phantom/phantom && brew install ashlrai/phantom/phantom
↗ Visit site★ GitHub
01

Features

01Reduces AI agent credential exposure risk by isolating secrets from agent context.
02Quick and easy local setup without requiring an account or custom CA.
03Native integrations with popular AI coding agents like Claude Code, Cursor, and Codex.
04Secure encrypted vault leveraging OS credential stores or an encrypted-file fallback.
05Tamper-evident audit log for security monitoring and integrity verification.
02

Compatibility

macOS
Runtime
Verified via docs
Linux
Runtime
Verified via docs
Windows
Runtime
Verified via docs
Rust
Built With
Verified via docs
Python
Runtime
Verified via docs
Claude Code
AI Client
Verified via docs
03

Quick start

1
$ brew tap ashlrai/phantom
2
$ brew trust --formula ashlrai/phantom/phantom
3
$ brew install ashlrai/phantom/phantom
04

Use cases

↳Securing API keys and sensitive credentials for AI coding agents in development workflows.
↳Preventing credential leaks in agent contexts, transcripts, tool calls, and generated files.
↳Local development of AI-powered applications requiring secure access to external APIs.
↳Facilitating secure team collaboration on projects with shared secrets and AI agent involvement.
05

Alternatives

⚡Top VS:
VSAgentHound★452→VSauthserver★81→VSsatgate★11→

Comments

Log in to leave a comment

No comments yet. Be the first!

On this page
01Features02Compatibility03Quick start04Use cases05Alternatives
VS
centian
★15
→
AgentHound logo
AgentHound★ 452
Offensive security framework for AI agent infrastructure - recon, credential looting, model exfiltration, poisoning, and attack-path analysis across MCP, A2A, gateways, and AI services. BloodHound for the agentic stack.
VSCompare→
authserver logo
authserver★ 81
OAuth 2.1 Authorization Server for the Model Context Protocol (MCP)
VSCompare→
satgate logo
satgate★ 11
The Economic Firewall for AI Agent Requests
VSCompare→
centian logo
centian★ 15
Give your AI agents structure, guardrails, and full observability — the Agent control plane built on MCP.
VSCompare→
shellgate logo
shellgate★ 15
Self hosted secure gateway for AI agents. One token. Full control. Complete audit trail.
VSCompare→
heddle logo
heddle★ 14
Heddle — The policy-and-trust layer for MCP tool servers. Turn YAML configs into validated, policy-enforced MCP tools.
VSCompare→
assay logo
assay★ 13
Open evidence profile for MCP tool actions. Enforce configured tool-call policies and record decisions and observations in offline-verifiable bundles. Optional Linux kernel controls. No hosted backend required.
VSCompare→
hol-guard logo
hol-guard★ 969
AI antivirus for developer agents: protect Codex, Claude Code, Cursor, Gemini, OpenCode, plugins, skills, MCP servers, and AI harnesses before tools run.
VSCompare→
See all alternatives →
ashlrai /
phantom-secrets
↗

Stop AI coding agents from leaking your API keys. Local proxy + MCP that swaps real secrets for phm_ tokens — works with Claude Code, Cursor, Windsurf, and Codex.

18MIT
Last commit4d ago
StatusActive
CategorySecurity & Safety
Links
Documentation↗Discussion↗Issues↗Releases↗
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit