Works
Only older checks are saved; no current product test.
Answers apply to each recorded setup, revision and operation. Earlier observations remain in Run details.
Tested on Linux| Platform | Haiku | Luna |
|---|
| Linux | yes | |
|---|
Linux / Haiku 5.5: yes · Oct 17, 2026 · 29b2871151b0 · claude-code / 2.1.300 (Claude Code)
Recorded operation: /workspace/target/release/ai-jail --no-agent-state --dry-run bash -c 'echo hi'
- Needs keys or browser
- None seen
- Setup time
- 2 min 4 s to install, 8 s to first use
- Issue
- none found in recorded check
Safety: no observations were recorded for the selected revision. Functional results do not establish a safety result.
Repair history and test corrections
Current diagnosis: unknown · resolvedsuperseded; repo akitaonrails/ai-jail has a later non-harness result a recorded run
Repair record fix-cd3ac45f83df73cecc7b · current status JSON. Original run outcomes remain unchanged.
Failed run: akitaonrails/ai-jail · Oct 16, 2026 · Our test failed; no repo verdict.
Test-repair rerun: akitaonrails/ai-jail · Oct 17, 2026 · Checked operation passed.
Current diagnosis: unknown · resolvedsuperseded; repo akitaonrails/ai-jail has a later non-harness result a recorded run
Repair record fix-9b3f3d0c9177ee6a5b36 · current status JSON. Original run outcomes remain unchanged.
Failed run: akitaonrails/ai-jail · Oct 16, 2026 · Our test failed; no repo verdict.
Repair development: akitaonrails/ai-jail · Oct 16, 2026 · Repair work (development) · not an initial-trial result.
Test-repair rerun: akitaonrails/ai-jail · Oct 17, 2026 · Checked operation passed.
Run details
Haiku 5.5
Native agent · Checked operation passedChecked operation /workspace/target/release/ai-jail --no-agent-state --dry-run bash -c 'echo hi'
Agent-reported scope: Built ai-jail 2.10.0 from source per README 'Build from source with Rust 1.97.1' and ran 'ai-jail --no-agent-state --dry-run bash -c echo hi', which prints the bubblewrap command line (/usr/bin/bwrap ...) without launching a sandbox.
Run details
- run
- akitaonrails--ai-jail-claude-haiku-5-5-1792211998270311452
- date
- Oct 17, 2026
- time to working
- 166.8s
- product hosting / execution
- Local command in the guest
- recorded operation scope
- product operation
- model
- claude-haiku-5-5
- agent / version
- claude-code / 2.1.300 (Claude Code)
- platform
- linux
- reported cost
- $0.005421
- Recorded source
- https://github.com/akitaonrails/ai-jail
- Tool transport
- mcp-assigned-vm
- Test version
- readme-3691798994c35a4248db43f18225b5f420ad493d7ca35c5ea2e8f1d0d89e7c20
- revision
- 29b2871151b0d0757a4a79199893f315932b7f14
Whose fault: not assigned · resolved What the agent did
Built ai-jail 2.10.0 from source per README 'Build from source with Rust 1.97.1' and ran 'ai-jail --no-agent-state --dry-run bash -c echo hi', which prints the bubblewrap command line (/usr/bin/bwrap ...) without launching a sandbox.
apt-get install -y bubblewrap build-essential pkg-config curl ca-certificates (README: Linux requires bwrap)
curl https://sh.rustup.rs | sh -s -- -y --profile minimal --default-toolchain 1.97.1 (README: Build from source with Rust 1.97.1; rust-toolchain.toml channel 1.97.1)
cargo build --release --locked (README: Build from source)
./target/release/ai-jail --version
./target/release/ai-jail --no-agent-state --dry-run bash -c 'echo hi' (README: Quick start, ai-jail --dry-run)
run JSON · checker evidence · agent report + check · transcript excerpt · error excerpt
No product verdict. Run record.
No product verdict. Run record.
No product verdict. Run record.