openguardrails/openguardrails
C3The vendor-neutral protocol for AI agent safety & security — and the neutral benchmark that ranks the vendors.
★ 50+ · openguardrails/openguardrails source on GitHub · this plugin in the registry
openguardrails/openguardrails is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It patches the dsh runtime, gates tool execution, executes system commands, starts a network server.
Installable plugin — declares a dsh.bundle manifest
What it can do
| Capability | Flag | Evidence |
|---|---|---|
| patches the dsh runtime | runtime_patch | ./cordis.patch.yml |
| gates tool execution | tool_gate | hook: tools/pre-execute |
| executes system commands | exec | ×6 in authored code, e.g. integrations/agent/claude-code/hooks/ogr-local.mjs:1344, integrations/agent/claude-code/hooks/ogr-local.mjs:1383 |
| starts a network server | net_server | ×3 in authored code, e.g. integrations/agent/claude-code/hooks/ogr-local.mjs:1569, integrations/agent/codex/hooks/ogr-local.mjs:1569 |
| reads credential-class env vars | token_env | OGR_API_KEY, OGR_ENROLL_TOKEN |
Services it injects
tools
Hooks it attaches
approval/request llm/stream tools/pre-execute tools/result
Outbound domains
openguardrails.com api.example.com api.anthropic.com chatgpt.com api.openai.com
Environment variables it reads
OGR_API_KEY OGR_LOCAL_PORT OGR_LOCAL_REDACTION OGR_RUNTIME_URL OGR_AGENT_WORKSPACE OGR_AGENT_USER OGR_TIMEOUT_MS OGR_AGENT_ID OGR_AGENT_TYPE OGR_LOCAL_UPSTREAM
How to read this
Levels measure capability surface and transparency, not maliciousness. A C3 plugin can be entirely legitimate — a desktop shell genuinely needs subprocesses. The point is that you can see this before installing. See the levels explained and how dsh plugins work.
Findings come from static analysis of shipped code; nothing is executed. Think a flag is wrong? Open an issue — every flag cites the file and line it came from.