中文
~/home / TypeScript / GitHub
GitHub · TypeScript

phantom

phantom is a workflow AI agent for Enable an AI to autonomously perform routine development tasks—like code commits, email replies, and credential management—while maintaining context across sessions. Pricing: Not specified. As of 2026-09-12, KanonAgent records 1.5k upvotes. First indexed by KanonAgent on 2026-07-24.
phantom is an AI co-worker with persistent memory and independent task execution, built on the Claude Agent SDK to handle routine development tasks like code commits and email replies.
1.5k upvotes
Tracked by Kanon since Jul 24, 2026
Reactivated
status
32/100
momentum · conf 0.83
20d
tracked since 2026-08-22
2026-09-11
last meaningful change
6
evidence records

Status basis: github_stars did not move for 7d, then gained 3 in the last 3d

📈 Timelinewhat changed, and when
🔎 Known / Unknown 11 of 23 fields unknown
Machine-callable— unknown —
Open source1 verified 1 quote(s)
Self-hostable1 inferred 1 quote(s)
Bring your own key— unknown —
Autonomy level2 inferred 2 quote(s)
Pricing model— unknown —
Integrationsslack,mcp inferred 2 quote(s)

“Unknown” means we have not verified it — it is not a “no”. Hard filters never treat unknown as false.

🤖 Agent teardown · workflow
Job to be doneEnable an AI to autonomously perform routine development tasks—like code commits, email replies, and credential management—while maintaining context across sessions.
AutonomyL2 · tool-calling(evidence: "installs software, spins up databases, builds dashboards…")
Who it is forDevelopers and remote teams
Prerequisitesopen source · self-hostable
Integrationsslack · mcp
PricingNot specified
Traction · why it is risingProductHunt votes: 1450, high community interest
Why it matters

Its persistent memory and secure credential handling represent a key step toward true AI co-pilots, with design patterns that could become industry standards.

Evidence quotesverbatim, from the product’s own materials

“public GitHub repository (README fetched)”— structural
“give the AI its own computer”— readme
“installs software, spins up databases, builds dashboards”— readme
“remembers what you told it last week, and gets measurably better”— readme
“MCP server, secure credential collection, email identity”— description
“runs on Slack”— readme
TypeScript
Signal source: GitHub
Visit official site →
📛 Official badgefor your site / README
phantom badge
Building phantom? Pick a style above — the embed code updates live. Deep color control via URL params: bg= / fg= / accent= (hex). It links back to this page.
Share on X
On mobile tap Share for WeChat / RED (Xiaohongshu) / X; on desktop use Copy text and paste into the app.

FAQ

What is phantom?

phantom is an AI co-worker with persistent memory and independent task execution, built on the Claude Agent SDK to handle routine development tasks like code commits and email replies.

What does phantom do?

Enable an AI to autonomously perform routine development tasks—like code commits, email replies, and credential management—while maintaining context across sessions.

Why does phantom matter?

Its persistent memory and secure credential handling represent a key step toward true AI co-pilots, with design patterns that could become industry standards.

How much does phantom cost?

Not specified

Is phantom open source?

Yes — phantom is open source.

Can I self-host phantom?

Yes — phantom can be self-hosted.

What does phantom integrate with?

slack,mcp

How popular is phantom?

As tracked by KanonAgent: 1.5k upvotes (first indexed 2026-07-24).

What are the best phantom alternatives?

Similar AI agents tracked by KanonAgent: langchain, BIG, claude-mem, AEO Engine, autogen, POST BRIDGE.

phantom alternatives — similar AI agents

langchainBIGclaude-memAEO EngineautogenPOST BRIDGE

Where this fits — browse the same shelf

AI Workflow Automation AgentsAI agents for Long-term memory and contextAI agents that work with SlackAI agents that work with MCP (Model Context Protocol)Self-hosted & open-source AI agents