A verification boundary for autonomous agents. Define outcomes, not procedures.
One agent, one prompt, full autonomy. Forge doesn't tell the agent what to do — it verifies whether the outcome was met. Verification is external, objective, and automatic.
$ forge run --spec specs/power-ups.md "implement power-ups"
# 8 power-ups implemented, verification passed — $6.03, ~8 minBun >= 1.2.0
curl -fsSL https://bun.sh/install | bash# From npm
bun install -g @vieko/forge
# From source
git clone https://github.com/vieko/forge.git
cd forge
bun install
bun run build
bun link # Makes `forge` available globallyforge run "implement feature X" # Run a task
forge run --spec specs/feature.md "implement this" # From a spec file
forge run --spec-dir ./specs/ -P "implement these" # Parallel specs
forge run --pending -P "implement pending" # Run only pending specs
forge run --resume <session-id> "continue" # Resume sessionforge define "build auth system" # Generate specs from description
forge audit specs/ # Audit codebase against specs
forge audit specs/ --fix # Audit-fix convergence loop
forge proof specs/feature.md # Generate test protocols
forge verify .forge/proofs/ # Run tests and create PR
forge pipeline "build auth system" # Full pipeline (define->run->audit->proof->verify)
forge tui # Interactive TUI
forge review # Review git changes
forge watch # Live-tail session logs
forge status # View recent results
forge stats # Aggregate run statistics
forge specs # List tracked specs with status
forge specs --add # Register all untracked specs
forge specs --resolve game.md # Mark spec as passed
forge specs --check # Auto-resolve implemented specsSee forge --help or forge <command> --help for all options.
- Prompt — wraps your task in an outcome-focused template
- Agent — single autonomous call, no orchestration
- Verify — auto-detects project type, runs build/test, feeds errors back (up to 3 attempts)
- Save — results, session logs, and cost to
.forge/results/
Specs can declare dependencies via frontmatter (depends: [a.md, b.md]) for ordered execution. Parallel runs use auto-tuned concurrency. A manifest (.forge/specs.json) tracks every spec from registration through execution. Failed specs can be rerun. Pending specs can be run selectively. Manually completed specs can be resolved. Sessions can be resumed or forked. Destructive commands are blocked. Transient errors retry automatically.
export ANTHROPIC_API_KEY=sk-ant-...Project-level overrides in .forge/config.json:
{ "maxTurns": 100, "maxBudgetUsd": 10.00 }- Bonfire — Session context persistence. Use
/bonfire specto create specs, then run them with Forge.
Forge exposes an MCP server for integration with Claude Code:
claude mcp add forge --scope user -t stdio -- bun /path/to/forge/dist/mcp.jsbun run src/index.ts run "test task" # Dev mode
bun run typecheck # Type check
bun run build # Build
bun test # Run testsAnimation by Jon Romero Ruiz.
MIT