Installation and Quickstart¶
AgentGuard supports Python 3.9, 3.10, 3.11, and 3.12. The product name is
AgentGuard, the production PyPI distribution is agentguard-evals, and the
Python import and terminal command remain agentguard.
Install from production PyPI¶
python -m pip install agentguard-evals
python -c "import agentguard; print(agentguard.__version__)"
agentguard --version
agentguard --help
For an isolated command installation:
Do not use TestPyPI as an installation source. Its similarly named project is unrelated to AgentGuard.
Initialize an existing project¶
agentguard init is included in the production
agentguard-evals==0.3.0 package. Preview project onboarding before writing
files:
Continue with safe project initialization for the generated-file inventory, conservative Python, Node.js, and Go detection rules, overwrite model, security boundaries, and first local and CI runs.
Run a deterministic safe evaluation¶
The ordinary PyPI package intentionally excludes repository examples. Clone the repository when you need benchmark fixtures, demo scripts, or committed result evidence:
git clone https://github.com/richinmrudul/agentguard.git
cd agentguard
python -m venv .venv
source .venv/bin/activate
python -m pip install -e ".[dev]"
agentguard run examples/configs/fix_auth_bug_local_command_safe.yaml --agent local-command
agentguard reports show --latest --type run
The fixture is deterministic, local, and network-free. It writes reports,
command evidence, a manifest, and a trace beneath .agentguard/, which is
ignored by Git.
Run the curated six-scenario showcase with:
Run the broader metadata-backed adversarial foundation with:
agentguard suite examples/suites/adversarial_core.yaml --allow-failures
.venv/bin/python scripts/adversarial_metrics.py --check
--allow-failures is intentional: every scenario in the current
adversarial-core suite represents expected unsafe behavior.
Choose an execution boundary¶
custom-commanduses the configured Docker execution path. Docker is required only for Docker-backed evaluations.local-commandandagent-commandexecute with host-user permissions and are not inherently sandboxed.- AgentGuard observes configured command/filesystem evidence and applies post-hoc policy checks. It is not syscall-level containment or a universal security boundary.
Continue with the demo, benchmark catalog, or real-agent evaluation guide.