guard-spec / guard-spec
Self-hosted prompt-injection firewall for LLM apps. OpenAI-compatible gateway with rules, an ONNX classifier and embeddings, output guard, alerts, SSO. English and Vietnamese.
Why this job: says: prompt injection; topics: ai-security, llm-security, guardrails, prompt-injection
What it does
- Three detection layers fused into one score. L1 rules (Vietnamese
- Indirect injection. tool results, retrieved documents and MCP tool
- Output guard. exfiltration links and images, personal data (including
- OpenAI-compatible (chat completions, the Responses API and embeddings), and
- Per-organization and per-project policies, written as text or in a visual
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit today
- Stars
- 10
- +10 this week
- Licence
- Apache-2.0
- Allows business use
- Downloads for
- 2 systems
- Mac, Linux
- Latest version
- v0.4.2
- Released today
- Built with
- Go
- 57% of the code
Is it looked after?
- Contributors
- 1
- Last 90 days
- Issues
- 100% closed
- 0 still open
- Pull requests
- 109 merged
- 20 waiting
- Releases
- 11
- 10 in the last year
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| uber / ADR | 2,128 | 24 days ago | Apache-2.0 | No downloads |
| superagent-ai / superagent | 6,855 | 2 months ago | MIT | Mac, Windows, Linux |
| dodoguardai / dodoguard | 276 | 56 days ago | MIT | No downloads |
More: download files, languages, topics and project record
Download files, v0.4.2
- Mac4 files, 4 downloads
- Linux8 files, 8 downloads
Languages
- Go 57%
- TypeScript 27%
- Python 11%
- CSS 1%
Topics
- ai-security
- cel
- golang
- guardrails
- jailbreak-detection
- llm-gateway
- llm-security
- mcp
- onnx
- openai-api
- prompt-injection
- rag
Project record
- JobAI security
- StatusNew to our list
- Commits346
- Forks0
- Watching0
- CreatedNov 3, 2026
- First seen by usNov 7, 2026
- CheckedNov 9, 2026