+
The dark software factory
for expert engineers
@@ -90,8 +92,8 @@ const cssExample = `/* All nodes default to fast + ch
and intervene only where it matters.
-
- $ curl -fsSL https://fabro.sh/install.sh | bash
+
+ $ curl -fsSL https://fabro.sh/install.sh | bash
-
+
-
+
+
+
+
-
-
-
-
+
+
+
+
Stop babysitting agents.
Start orchestrating workflows.
@@ -133,116 +138,102 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
- Why teams use Fabro
-
-
-
- →
-
- Extend disengagement time
- Define a workflow with verification gates and walk away. Fabro keeps the process on track without you.
-
-
-
- →
-
- Leverage ensemble intelligence
- Use one model to implement, another to cross-critique, and a third to summarize — all in a single workflow.
-
-
-
- →
-
- Guarantee code quality
- Layer test suites, linters, type checkers, and LLM-as-judge into your workflow graph. Failures trigger fix loops automatically.
-
-
-
- →
-
- Reduce token bills
- Route cheap tasks to fast models and reserve frontier models for the steps that need them. One-line stylesheet change.
-
-
-
- →
-
- Improve agent security
- Run agents in cloud sandboxes with full network and filesystem isolation. Keep untrusted code off your laptop.
-
-
-
- →
-
- Run agents 24/7
- The API server queues and executes runs continuously. Close your laptop — results are waiting when you return.
-
-
-
-
-
-
-
-
-
-
+
+
+
Key features
+
- 🔀
+
+
+
+ Workflow Engine
Deterministic workflow graphs
Define pipelines in Graphviz DOT with branching, loops, parallelism, and human gates. Diffable, reviewable, version-controlled.
+
- 🙋
+
+
+
+ Control
Human-in-the-loop
Approval gates pause for human decisions. Steer running agents mid-turn. Interview steps collect structured input.
+
- 🎨
+
+
+
+ Intelligence
Multi-model routing
CSS-like stylesheets route each node to the right model and provider, with automatic fallback chains.
+
- ☁️
+
+
+
+ Isolation
Cloud sandboxes
Run agents in isolated cloud VMs with snapshot-based setup, network controls, and automatic cleanup. SSH in with fabro ssh.
+
- 🌲
+
+
+
+ Versioning
Git checkpointing
Every stage commits code changes and execution metadata to Git branches. Resume, revert, or trace any change.
+
- 📊
+
+
+
+ Analytics
Automatic retrospectives
Each run generates a retro with cost, duration, files touched, and an LLM-written narrative. Your workflows improve over time.
@@ -252,11 +243,19 @@ const cssExample = `/* All nodes default to fast + ch
+
- ⚡
+
+
+
+ Interface
API server + web UI
REST API with SSE event streaming and a React web UI. Run workflows programmatically or as a service. Queue runs, track progress, manage at scale.
@@ -264,11 +263,18 @@ const cssExample = `/* All nodes default to fast + ch
+
- 🦀
+
+
+
+ Distribution
Single binary, no runtime
One compiled Rust executable with zero dependencies. No Python, no Node, no Docker required.
@@ -281,12 +287,11 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
+
+
-
-
+
+
Workflow-as-Code
@@ -313,10 +318,56 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
+
+
+
+
@@ -335,17 +386,19 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
+
+
-
+
-
+
+
+
+
-
-
+
+
Verification is a
first-class concept
@@ -360,13 +413,14 @@ const cssExample = `/* All nodes default to fast + ch
-
+
-
+
+
-
-
+
+
Multi-model, multi-provider
@@ -389,7 +443,7 @@ const cssExample = `/* All nodes default to fast + ch
-
+
@@ -399,124 +453,67 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
-
-
-
-
-
- $
- fabro run deploy --sandbox daytona
-
-
- Provisioning sandbox... done (3.2s)
- Cloning repository... done
- Running workflow... stage 1/5
-
-
- $
- fabro ssh
- ← shell into the running sandbox
-
-
- $
- fabro preview 3000
- ← expose a port for live debugging
-
-
-
-
-
-
- Cloud sandboxes with
SSH and preview links
+
+
+
+
+
+
+
+ Ground-up observability
-
- Run agents in isolated cloud VMs — not on your laptop. Full network and filesystem isolation,
- snapshot-based setup, and automatic cleanup.
+
+ Full traceability across runs — every model call, every tool invocation, every decision point.
+ Query run data with SQL. Automatic retrospectives rate each run and surface friction points.
-
- -
-
- Shell into running sandboxes with
fabro ssh
-
- -
-
- Expose ports with
fabro preview for live debugging
-
- -
-
- 6 providers: local, Docker, Daytona, SSH, exe.dev, Sprites
-
-
-
-
-
-
-
-
-
-
-
- Ground-up observability
-
-
- Full traceability across runs — every model call, every tool invocation, every decision point.
- Query run data with SQL. Automatic retrospectives rate each run and surface friction points.
-
-
-
-
-
-
-
- fabro.workflow.run — trace
-
-
-
- workflow
-
-
- fabro.workflow.run — 24.3s
+
+
+
+
+ fabro.workflow.run — trace
+
+
+
+ workflow
+
+
-
-
- plan
-
-
- 4.1s
+
+ plan
+
+
-
-
- implement
-
-
- 12.7s
+
+ implement
+
+
-
-
- simplify
-
-
- 4.2s
+
+ simplify
+
+
-
-
- retro
-
-
- 3.3s
+
+ retro
+
+
@@ -525,56 +522,18 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
- Powerful CLI
-
-
- One binary. Zero dependencies. Everything you need to run, inspect, and manage workflows.
-
-
-
-
-
- Run workflows
- $ fabro run plan-implement
- $ fabro run --sandbox daytona --resume
-
-
- Manage runs
- $ fabro ps
- $ fabro logs <run-id>
- $ fabro diff <run-id>
-
-
- Debug sandboxes
- $ fabro ssh
- $ fabro preview 3000
-
-
- Ship results
- $ fabro pr create <run-id>
-
-
-
-
-
-
-
-
-
-
-
- Shippable outcomes,
not impressive output
+
+
+
+
+
+
+ Workflows you can audit.
+ Output you can trust.
-
- Most AI coding tools optimize for impressive output. Fabro optimizes for what actually merges.
-
-
+
+
+
→
@@ -604,22 +563,10 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
-
-
-
- Workflows you can audit.
- Output you can trust.
-
-
-
- $ curl -fsSL https://fabro.sh/install.sh | bash
+
+
+ $ curl -fsSL https://fabro.sh/install.sh | bash
for expert engineers
$ curl -fsSL https://fabro.sh/install.sh | bash
+ $ curl -fsSL https://fabro.sh/install.sh | bash
+
+
+
+
+
+
Stop babysitting agents.
Start orchestrating workflows.
@@ -133,116 +138,102 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
- Why teams use Fabro
-
-
-
- →
-
- Extend disengagement time
- Define a workflow with verification gates and walk away. Fabro keeps the process on track without you.
-
-
-
- →
-
- Leverage ensemble intelligence
- Use one model to implement, another to cross-critique, and a third to summarize — all in a single workflow.
-
-
-
- →
-
- Guarantee code quality
- Layer test suites, linters, type checkers, and LLM-as-judge into your workflow graph. Failures trigger fix loops automatically.
-
-
-
- →
-
- Reduce token bills
- Route cheap tasks to fast models and reserve frontier models for the steps that need them. One-line stylesheet change.
-
-
-
- →
-
- Improve agent security
- Run agents in cloud sandboxes with full network and filesystem isolation. Keep untrusted code off your laptop.
-
-
-
- →
-
- Run agents 24/7
- The API server queues and executes runs continuously. Close your laptop — results are waiting when you return.
-
-
-
-
-
-
-
-
-
-
+
+
+
Key features
+
- 🔀
+
+
+
+ Workflow Engine
Deterministic workflow graphs
Define pipelines in Graphviz DOT with branching, loops, parallelism, and human gates. Diffable, reviewable, version-controlled.
+
- 🙋
+
+
+
+ Control
Human-in-the-loop
Approval gates pause for human decisions. Steer running agents mid-turn. Interview steps collect structured input.
+
- 🎨
+
+
+
+ Intelligence
Multi-model routing
CSS-like stylesheets route each node to the right model and provider, with automatic fallback chains.
+
- ☁️
+
+
+
+ Isolation
Cloud sandboxes
Run agents in isolated cloud VMs with snapshot-based setup, network controls, and automatic cleanup. SSH in with fabro ssh.
+
- 🌲
+
+
+
+ Versioning
Git checkpointing
Every stage commits code changes and execution metadata to Git branches. Resume, revert, or trace any change.
+
- 📊
+
+
+
+ Analytics
Automatic retrospectives
Each run generates a retro with cost, duration, files touched, and an LLM-written narrative. Your workflows improve over time.
@@ -252,11 +243,19 @@ const cssExample = `/* All nodes default to fast + ch
+
- ⚡
+
+
+
+ Interface
API server + web UI
REST API with SSE event streaming and a React web UI. Run workflows programmatically or as a service. Queue runs, track progress, manage at scale.
@@ -264,11 +263,18 @@ const cssExample = `/* All nodes default to fast + ch
+
- 🦀
+
+
+
+ Distribution
Single binary, no runtime
One compiled Rust executable with zero dependencies. No Python, no Node, no Docker required.
@@ -281,12 +287,11 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
+
+
-
-
+
+
Workflow-as-Code
@@ -313,10 +318,56 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
+
+
+
+
@@ -335,17 +386,19 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
+
+
-
+
-
+
+
+
+
-
-
+
+
Verification is a
first-class concept
@@ -360,13 +413,14 @@ const cssExample = `/* All nodes default to fast + ch
-
+
-
+
+
-
-
+
+
Multi-model, multi-provider
@@ -389,7 +443,7 @@ const cssExample = `/* All nodes default to fast + ch
-
+
@@ -399,124 +453,67 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
-
-
-
-
-
- $
- fabro run deploy --sandbox daytona
-
-
- Provisioning sandbox... done (3.2s)
- Cloning repository... done
- Running workflow... stage 1/5
-
-
- $
- fabro ssh
- ← shell into the running sandbox
-
-
- $
- fabro preview 3000
- ← expose a port for live debugging
-
-
-
-
-
-
- Cloud sandboxes with
SSH and preview links
+
+
+
+
+
+
+
+ Ground-up observability
-
- Run agents in isolated cloud VMs — not on your laptop. Full network and filesystem isolation,
- snapshot-based setup, and automatic cleanup.
+
+ Full traceability across runs — every model call, every tool invocation, every decision point.
+ Query run data with SQL. Automatic retrospectives rate each run and surface friction points.
-
- -
-
- Shell into running sandboxes with
fabro ssh
-
- -
-
- Expose ports with
fabro preview for live debugging
-
- -
-
- 6 providers: local, Docker, Daytona, SSH, exe.dev, Sprites
-
-
-
-
-
-
-
-
-
-
-
- Ground-up observability
-
-
- Full traceability across runs — every model call, every tool invocation, every decision point.
- Query run data with SQL. Automatic retrospectives rate each run and surface friction points.
-
-
-
-
-
-
-
- fabro.workflow.run — trace
-
-
-
- workflow
-
-
- fabro.workflow.run — 24.3s
+
+
+
+
+ fabro.workflow.run — trace
+
+
+
+ workflow
+
+
-
-
- plan
-
-
- 4.1s
+
+ plan
+
+
-
-
- implement
-
-
- 12.7s
+
+ implement
+
+
-
-
- simplify
-
-
- 4.2s
+
+ simplify
+
+
-
-
- retro
-
-
- 3.3s
+
+ retro
+
+
@@ -525,56 +522,18 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
- Powerful CLI
-
-
- One binary. Zero dependencies. Everything you need to run, inspect, and manage workflows.
-
-
-
-
-
- Run workflows
- $ fabro run plan-implement
- $ fabro run --sandbox daytona --resume
-
-
- Manage runs
- $ fabro ps
- $ fabro logs <run-id>
- $ fabro diff <run-id>
-
-
- Debug sandboxes
- $ fabro ssh
- $ fabro preview 3000
-
-
- Ship results
- $ fabro pr create <run-id>
-
-
-
-
-
-
-
-
-
-
-
- Shippable outcomes,
not impressive output
+
+
+
+
+
+
+ Workflows you can audit.
+ Output you can trust.
-
- Most AI coding tools optimize for impressive output. Fabro optimizes for what actually merges.
-
-
+
+
+
→
@@ -604,22 +563,10 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
-
-
-
- Workflows you can audit.
- Output you can trust.
-
-
-
- $ curl -fsSL https://fabro.sh/install.sh | bash
+
+
+ $ curl -fsSL https://fabro.sh/install.sh | bash
+
Stop babysitting agents.
Start orchestrating workflows.
@@ -133,116 +138,102 @@ const cssExample = `/* All nodes default to fast + ch
- Why teams use Fabro -
-
+
+
+
Key features
+
- 🔀
+
+
+
+ Workflow Engine
Deterministic workflow graphs
Define pipelines in Graphviz DOT with branching, loops, parallelism, and human gates. Diffable, reviewable, version-controlled.
+
- 🙋
+
+
+
+ Control
Human-in-the-loop
Approval gates pause for human decisions. Steer running agents mid-turn. Interview steps collect structured input.
+
- 🎨
+
+
+
+ Intelligence
Multi-model routing
CSS-like stylesheets route each node to the right model and provider, with automatic fallback chains.
+
- ☁️
+
+
+
+ Isolation
Cloud sandboxes
Run agents in isolated cloud VMs with snapshot-based setup, network controls, and automatic cleanup. SSH in with fabro ssh.
+
- 🌲
+
+
+
+ Versioning
Git checkpointing
Every stage commits code changes and execution metadata to Git branches. Resume, revert, or trace any change.
+
- 📊
+
+
+
+ Analytics
Automatic retrospectives
Each run generates a retro with cost, duration, files touched, and an LLM-written narrative. Your workflows improve over time.
@@ -252,11 +243,19 @@ const cssExample = `/* All nodes default to fast + ch
+
- ⚡
+
+
+
+ Interface
API server + web UI
REST API with SSE event streaming and a React web UI. Run workflows programmatically or as a service. Queue runs, track progress, manage at scale.
@@ -264,11 +263,18 @@ const cssExample = `/* All nodes default to fast + ch
+
- 🦀
+
+
+
+ Distribution
Single binary, no runtime
One compiled Rust executable with zero dependencies. No Python, no Node, no Docker required.
@@ -281,12 +287,11 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
+
+
-
-
+
+
Workflow-as-Code
@@ -313,10 +318,56 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
+
+
+
+
@@ -335,17 +386,19 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
+
+
-
+
-
+
+
+
+
-
-
+
+
Verification is a
first-class concept
@@ -360,13 +413,14 @@ const cssExample = `/* All nodes default to fast + ch
-
+
-
+
+
-
-
+
+
Multi-model, multi-provider
@@ -389,7 +443,7 @@ const cssExample = `/* All nodes default to fast + ch
-
+
@@ -399,124 +453,67 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
-
-
-
-
-
- $
- fabro run deploy --sandbox daytona
-
-
- Provisioning sandbox... done (3.2s)
- Cloning repository... done
- Running workflow... stage 1/5
-
-
- $
- fabro ssh
- ← shell into the running sandbox
-
-
- $
- fabro preview 3000
- ← expose a port for live debugging
-
-
-
-
-
-
- Cloud sandboxes with
SSH and preview links
+
+
+
+
+
+
+
+ Ground-up observability
-
- Run agents in isolated cloud VMs — not on your laptop. Full network and filesystem isolation,
- snapshot-based setup, and automatic cleanup.
+
+ Full traceability across runs — every model call, every tool invocation, every decision point.
+ Query run data with SQL. Automatic retrospectives rate each run and surface friction points.
-
- -
-
- Shell into running sandboxes with
fabro ssh
-
- -
-
- Expose ports with
fabro preview for live debugging
-
- -
-
- 6 providers: local, Docker, Daytona, SSH, exe.dev, Sprites
-
-
-
-
-
-
-
-
-
-
-
- Ground-up observability
-
-
- Full traceability across runs — every model call, every tool invocation, every decision point.
- Query run data with SQL. Automatic retrospectives rate each run and surface friction points.
-
-
-
-
-
-
-
- fabro.workflow.run — trace
-
-
-
- workflow
-
-
- fabro.workflow.run — 24.3s
+
+
+
+
+ fabro.workflow.run — trace
+
+
+
+ workflow
+
+
-
-
- plan
-
-
- 4.1s
+
+ plan
+
+
-
-
- implement
-
-
- 12.7s
+
+ implement
+
+
-
-
- simplify
-
-
- 4.2s
+
+ simplify
+
+
-
-
- retro
-
-
- 3.3s
+
+ retro
+
+
@@ -525,56 +522,18 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
- Powerful CLI
-
-
- One binary. Zero dependencies. Everything you need to run, inspect, and manage workflows.
-
-
-
-
-
- Run workflows
- $ fabro run plan-implement
- $ fabro run --sandbox daytona --resume
-
-
- Manage runs
- $ fabro ps
- $ fabro logs <run-id>
- $ fabro diff <run-id>
-
-
- Debug sandboxes
- $ fabro ssh
- $ fabro preview 3000
-
-
- Ship results
- $ fabro pr create <run-id>
-
-
-
-
-
-
-
-
-
-
-
- Shippable outcomes,
not impressive output
+
+
+
+
+
+
+ Workflows you can audit.
+ Output you can trust.
-
- Most AI coding tools optimize for impressive output. Fabro optimizes for what actually merges.
-
-
+
+
+
→
@@ -604,22 +563,10 @@ const cssExample = `/* All nodes default to fast + ch
-
-
-
-
-
-
-
-
-
- Workflows you can audit.
- Output you can trust.
-
-
-
- $ curl -fsSL https://fabro.sh/install.sh | bash
+
+
+ $ curl -fsSL https://fabro.sh/install.sh | bash
Key features
Workflow Engine
Deterministic workflow graphs
Define pipelines in Graphviz DOT with branching, loops, parallelism, and human gates. Diffable, reviewable, version-controlled.
Control
Human-in-the-loop
Approval gates pause for human decisions. Steer running agents mid-turn. Interview steps collect structured input.
Intelligence
Multi-model routing
CSS-like stylesheets route each node to the right model and provider, with automatic fallback chains.
Isolation
Cloud sandboxes
Run agents in isolated cloud VMs with snapshot-based setup, network controls, and automatic cleanup. SSH in with fabro ssh.
Versioning
Git checkpointing
Every stage commits code changes and execution metadata to Git branches. Resume, revert, or trace any change.
Analytics
Automatic retrospectives
Each run generates a retro with cost, duration, files touched, and an LLM-written narrative. Your workflows improve over time.
@@ -252,11 +243,19 @@ const cssExample = `/* All nodes default to fast + ch
Interface
REST API with SSE event streaming and a React web UI. Run workflows programmatically or as a service. Queue runs, track progress, manage at scale.
@@ -264,11 +263,18 @@ const cssExample = `/* All nodes default to fast + ch
Distribution
One compiled Rust executable with zero dependencies. No Python, no Node, no Docker required.
@@ -281,12 +287,11 @@ const cssExample = `/* All nodes default to fast + ch
-
@@ -313,10 +318,56 @@ const cssExample = `/* All nodes default to fast + ch
@@ -360,13 +413,14 @@ const cssExample = `/* All nodes default to fast + ch
@@ -389,7 +443,7 @@ const cssExample = `/* All nodes default to fast + ch
- Run agents in isolated cloud VMs — not on your laptop. Full network and filesystem isolation,
- snapshot-based setup, and automatic cleanup.
+
+ Full traceability across runs — every model call, every tool invocation, every decision point.
+ Query run data with SQL. Automatic retrospectives rate each run and surface friction points.
- Full traceability across runs — every model call, every tool invocation, every decision point.
- Query run data with SQL. Automatic retrospectives rate each run and surface friction points.
-
- One binary. Zero dependencies. Everything you need to run, inspect, and manage workflows.
-
- Most AI coding tools optimize for impressive output. Fabro optimizes for what actually merges.
- API server + web UI
Single binary, no runtime
+
Workflow-as-Code
+
+
+
+
Verification is a
first-class concept
+
Multi-model, multi-provider
- Cloud sandboxes with
SSH and preview links
+
+
+
+
+
+ Ground-up observability
-
-
fabro ssh
- fabro preview for live debugging
-
- Ground-up observability
-
-
- Powerful CLI
-
-
- Shippable outcomes,
not impressive output
+
+
+ Workflows you can audit.
-
+ Output you can trust.
- Workflows you can audit.
-
- Output you can trust.
- $ curl -fsSL https://fabro.sh/install.sh | bash
+ $ curl -fsSL https://fabro.sh/install.sh | bash