By matanryngler
Production safety guardrails that block write/mutating operations on cloud, database, IaC, and deployment CLIs.
Cloud infrastructure and production safety guardrails - prevents write/mutating operations on cloud, database, and deployment CLIs
Production safety guardrails - prevents write/mutating operations on cloud, database, and deployment CLIs
DeployShield is a cross-platform production safety guardrail for Claude Code and Gemini CLI. It intercepts terminal commands before execution and blocks dangerous operations (writes, deletes, etc.) while allowing read-only commands to pass through.
Large Language Models (LLMs) are incredibly capable but can accidentally execute destructive commands in production environments. DeployShield provides a deterministic safety layer that doesn't rely on probabilistic model instructions.
sudo, and bash -c.production while allowing them in dev.| Category | Guarded CLIs |
|---|---|
| ☁️ Cloud | aws, gcloud, az, kubectl, helm |
| 🗄️ Databases | psql, mysql, mongosh, redis-cli |
| 🏗️ IaC | terraform, pulumi, cdk, sam, serverless (sls), ansible-playbook |
| 📦 Publishing | npm, yarn, pnpm, cargo, twine, gem |
| 🔧 Other | vault, gh, docker, podman |
/plugin marketplace add matanryngler/deployshield
/plugin install deployshield
gemini extensions install https://github.com/matanryngler/deployshield
By default, DeployShield blocks ALL write operations. Create a .deployshield.json file to allow writes in non-production contexts.
.deployshield.json to your project repository to ensure that every developer follows the same safety standards.{
"kubectl": ["prod-cluster", "production", "prod-*"],
"aws": ["production-profile"],
"terraform": ["prod-workspace"]
}
For detailed configuration options, see the Configuration Guide.
DeployShield provides deep protection that handles common bypass attempts:
echo $(terraform destroy) → Blockedsudo kubectl delete ... → Blockedecho pod-id | xargs kubectl delete pod → Blockedbash -c "aws s3 rm ..." → Blockedcat <(pulumi destroy) → BlockedThis project uses uv for dependency management and pre-commit for quality control.
To understand how DeployShield works under the hood, check the Internals Guide.
# Run tests
uv run pytest -v
# Install pre-commit hooks
uv run pre-commit install
MIT
Executes bash commands
Hook triggers when Bash tool is used
No model invocation
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Executes directly as bash, bypassing the AI model
Executes directly as bash, bypassing the AI model
npx claudepluginhub matanryngler/deployshield --plugin deployshieldA secure runtime for Claude Code. Intercepts every tool call with policy-based allow/block/ask decisions, evasion detection, path fencing, file snapshots, and audit logging.
Runtime security plugin for Claude Code with balanced default hooks plus the Stallion inline MCP gateway for shell, git, MCP, secret, and exfiltration risks.
Protect secrets and credentials from Claude Code. Blocks writes to .env files, detects API keys in shell commands, prevents hardcoded tokens, and guards service account JSON files.
Pre-execution safety layer that blocks dangerous shell commands and credential file reads using pattern matching + LLM analysis. Fail-closed design.
Achieve flow state safely with Claude Code. Auto-approves routine work, gates risky actions, hard-blocks dangerous patterns. Dual enforcement (skill + hooks), token cap for cost governance, full audit trail. Zero dependencies.
Smart command safety filter for Claude Code — parses shell pipelines and evaluates per-command safety rules to auto-approve safe commands and block dangerous ones