Reusable / Secure / Adorable AI Agent
CLI plus embeddable Go core. Security presets, auditable execution, and human-friendly interaction.
[ START // AGENT ]
Command Line
Copy the live install path for shells, remote machines, and automation.
$ curl -fsSL -o /tmp/install.sh https://raw.githubusercontent.com/quailyquaily/mistermorph/refs/heads/master/scripts/install-release.sh
$ sudo bash /tmp/install.sh
$ _
Desktop App
Visual onboarding, session history, and multi-agent control in one place.
[ CAPABILITY // MATRIX ]
Every step is visible, testable, and auditable.
Keep the CLI experience, and embed the same core into your product.
Credentials, redaction, policy checks, and approvals are built in.
Loads SKILL.md context automatically when needed.
Gets context, speaks plainly, and keeps a human touch when it matters.
Supports OpenAI, Claude, Bedrock, and more.
[ SYSTEM // CATALOG ]
Execution safety chain
Prompt Injection Protection
Blocks malicious prompt overrides and constrains reachable targets.
Redaction Filter
Tool observations and final output are sanitized before exposure.
Approval + Audit
Risky actions pause for approval and emit audit events.
Aqua collaboration mesh
Translates 'I am fine with anything' into one plan nobody hates.
Compares everything and refuses to let you pay full price.
Reminds you about passport, charger, and obvious life choices.
Channel topology