phivsVT Code
Generated from the two spec rows. Green marks the better value where a spec has a clear direction. Everything else is just different.
phi
Pulse AI Club · Terminal agent
#28OSSMCP
Panel
6.92 spec wins
- Reliability
- 7.2
- Usefulness
- 6.5
- Cost
- 8.0
- Longevity
- 6.0
“Its extensions talk over a binary protocol on standard input and output, because JSON was apparently the slow part of asking a model.”
VT Code
vinhnx · Terminal agent
#60OSSMCP
Panel
7.03 spec wins
- Reliability
- 6.5
- Usefulness
- 6.7
- Cost
- 8.8
- Longevity
- 6.2
“It supports skills and a plugin system, so you can extend it in two ways and later discover which one is maintained.”
Spec by spec
| Spec | phi | VT Code |
|---|---|---|
| Architecture | ||
| Category | Terminal agent | Terminal agent |
| Runs | local | local |
| Platforms | macos, linux, windows | macos, linux |
| Context window | not documented | not documented |
| Protocols | ||
| MCP client | Yes | Yes |
| MCP server | No | No |
| Capabilities | ||
| Runs terminal commands | Yes | Yes |
| Multi-file edits | Yes | Yes |
| Git operations | No | Yes |
| Browser control | NoExternal HTTP fetch is available only through an MCP server you configure. | YesOnly through the opt-in WebMCP browser bridge. |
| Sandboxed execution | No | No |
| Multi-agent orchestration | Yes | No |
| Headless / CI mode | Yes"phi run -p" executes one agent loop headlessly, and a headless-strict mode is one of the configurable permission modes. | No |
| Models | ||
| Backbone | OpenAI-compatible, Anthropic | OpenAI, Ollama, LM Studio, llama.cpp |
| Bring your own model | Yes | Yes |
| Local models | YesAny OpenAI-compatible base URL can be configured, including a locally served endpoint. | Yes |
| Cost | ||
| Pricing model | byok | byok |
| Starts at | $0/mo | $0/mo |
| Free tier | Yes | Yes |
| Bring your own key | Yes | Yes |
| Openness | ||
| Open source | Yes | Yes |
| License | MIT | Apache-2.0 |
| GitHub stars | 524 | 860 |
Which one would each critic pick
| Critic | phi | VT Code | Pick |
|---|---|---|---|
| El Juez | — | — | not enough reviews |
| El Amigo | 7.3 | 7.5 | VT Code — Pick it if you want one agent that answers in the terminal and inside Zed; pick an editor-native tool if you never leave the editor. |
| El Crítico | 6.3 | 6.8 | VT Code — It runs shell commands and performs git operations, and the row records no isolation of any kind, so the boundary is whatever your account can reach. |
| El Profesor | 7.5 | 7.3 | phi — Sub-agents run isolated jobs whose turns stay out of the parent's context while remaining visible in the interface, which separates what the operator sees from what the model reads. |
| La Inversora | 5.5 | 6.5 | VT Code — 832 stars, one maintainer, no company and nothing hosted: a good tool in a category where good tools are the commodity. |
| La Jefa | 6.3 | 6.0 | phi — Free at any headcount, and it runs one loop non-interactively with a strict headless permission mode, which makes it the rare desktop tool that can also be a pipeline step. |
| El Hacker | 8.8 | 8.3 | phi — MIT, any compatible base URL including one I serve myself, and tool servers listed by name only until the agent calls list, inspect and invoke to reach them on demand. |
Picks are derived from each critic's own scores. Humans vote on matchups on the duels page.