sheeki03/tirith
Terminal security for developers and AI agents. Intercepts homograph URLs, pipe-to-shell, ANSI injection, obfuscated payloads, data exfiltration, and malicious AI skills/configs before they execute. observed · 2026-08-28
Health v2 · maintenance only
78/100
- Activity 99
- Release rhythm 87
- Longevity 15
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: 1
- age_days: 212
- days_rel: 8
- days_push: 7
- n_releases_24m: 86
Adoption not part of the score
2683 stars · 93 forks observed · 2026-08-28
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded
Tirith is a Rust-based terminal security tool that intercepts shell commands, pasted content, and files to block threats like homograph URLs, pipe-to-shell, ANSI injection, credential exfiltration, and malicious AI configs before they execute. It integrates via shell hooks, an MCP server for AI coding agents, CI pipelines, and an LSP server, backed by 200+ detection rules and a signed threat intelligence database.
Use cases
- block curl pipe-to-shell commands with homograph URLs before they execute
- protect my terminal from pasted commands containing hidden unicode or ANSI escape characters
- secure AI coding agents like Claude Code or Cursor from running malicious shell commands
- detect typosquatted git repos and malicious npm/pip packages before install
- scan CLAUDE.md and MCP configs for prompt injection and hidden instructions
- detect API keys and secrets being leaked in terminal commands
- add terminal command scanning to CI with SARIF output
When to choose
- you run shell commands from AI agents and want a fail-closed guard against prompt-injected or obfuscated commands
- you want browser-grade protection against homograph/punycode URLs in your terminal
- you need supply-chain checks for git, docker, pip, and npm commands
When to avoid
- you need a hardened authorization boundary - the docs note enforcement varies by shell and mode
- you require a permissive license - it is AGPL-3.0
- you need zero-latency guarantees on shells or terminals with untested hook support
Facets
cli-tool · maturity active
security cli developer-tools mcp vulnerability-scanning security developer-tools windows cli rust terminal-security homograph-attack shell-hook pipe-to-shell ansi-injection threat-intelligence ai-agent-security supply-chain-security unicode-security command-line ai-agents macos linux
7 sources
- readme: https://github.com/sheeki03/tirith · fetched 2026-08-28 · 45104bc3cbe1
- homepage: https://tirith.sh · fetched 2026-08-29 · 1c4599a2ceb9
- site_page: https://tirith.sh/docs · fetched 2026-08-29 · 7a5c03309f95
- site_page: https://tirith.sh/docs/detection-rules · fetched 2026-08-29 · 6ecc15a0b9aa
- site_page: https://tirith.sh/docs/mcp-setup · fetched 2026-08-29 · 57ac324df31a
- site_page: https://tirith.sh/docs/getting-started · fetched 2026-08-29 · 9338a5d00f8e
- registry_crates: https://crates.io/api/v1/crates/tirith · fetched 2026-08-29 · 601d03283490
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| sheeki03/tirith | main | 78 |
For agents
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem