Repository
meta-harness-tbench2-artifact
's Terminus-2 framework. 76.4% on Terminal-Bench 2.0 (89 tasks × 5 trials, Claude Opus 4.6). · License unknown · Install via pip · Maintenance unknown
- License
- Unknown
- Split
- N
- Easy
- 4
- Medium
- 55
Specs
- SplitN
- Easy4
- Medium55
- Hard30
README
Meta-Harness
Agent scaffold for Terminal-Bench 2.0, built on top of Terminus-KIRA by KRAFTON AI and Harbor's Terminus-2 framework.
Results
76.4% on Terminal-Bench 2.0 (89 tasks × 5 trials, Claude Opus 4.6).
| Split | N | Score |
|---|---|---|
| Easy | 4 | 100.0 |
| Medium | 55 | 81.1 |
| Hard | 30 | 64.7 |
| All | 89 | 76.4 |
Usage
pip install harbor
export ANTHROPIC_API_KEY=<your-key>
harbor run \
--agent-import-path agent:AgentHarness \
-d [email protected] \
-m anthropic/claude-opus-4-6 \
-e runloop \
-n 20 \
--n-attempts 5Method
Meta-Harness extends the Terminus-KIRA agent with environment bootstrapping: before the agent loop starts, it gathers a snapshot of the sandbox environment (working directory, file listing, available languages/tools, package managers, memory) and injects it into the initial prompt. This saves 2-5 early exploration turns that the agent normally spends on ls, which python3, etc.
The agent was discovered through automated harness evolution. More details coming soon.
Acknowledgements
We thank KRAFTON AI for compute support.