By sonomoshq
Sonomos Canary — persistent PII leak counter for Claude Code. Tracks every piece of sensitive data you expose to AI across sessions. Regex + Claude self-scan hybrid detection with a running counter and interactive dashboard.
View your PII exposure dashboard — shows all sensitive data detected across Claude Code sessions with category breakdown, timeline, and details. Use when the user asks about PII leaks, privacy exposure, or wants to see their Sonomos dashboard.
Deep-scan the full conversation history for PII. Automatic scanning covers only the most recent message — use this for thorough audits of the entire session. Detects names, addresses, legal IDs, medical records, trade secrets, crypto credentials, API tokens, and 70+ other semantic categories.
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
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.
This plugin requires configuration values that are prompted when the plugin is enabled. Sensitive values are stored in your system keychain.
llm_scan_enabledEnable automatic Claude self-scan for semantic PII after every prompt (default: true)
${user_config.llm_scan_enabled}confidence_thresholdMinimum confidence to record a detection: high or medium (default: medium)
${user_config.confidence_threshold}
You have no idea how much PII you've fed to Claude.
Canary is a privacy plugin for Claude Code that counts every piece of sensitive data you expose across all sessions.
Credit cards. SSNs. API keys. Emails. Medical records. Crypto wallets. Names. Addresses.
The number only goes up.
/plugin marketplace add sonomos-ai/Canary-Plugin
/plugin install canary@sonomos
No API keys. No external services. No config. Two commands and you're running.
|
16 Regex Detectors (every message, ~10ms) Real checksum validation, not just pattern matching:
|
70+ Semantic Categories (Claude self-scan) Claude scans its own context for PII that regex can't catch:
|
You type a message
|
v
Claude processes it ──> Stop hook fires (async, invisible)
|
┌─────────┴─────────┐
Regex Detectors Claude Self-Scan
(16 patterns + checksums) (70+ categories)
└─────────┬─────────┘
v
~/.sonomos/leaks.jsonl
|
Session start ──> counter displayed
| Command | What it does |
|---|---|
/canary:leaked | Open the interactive HTML dashboard |
/canary:leaked stats | Print a text summary |
/canary:scan | Deep-scan the full conversation history |
/canary:leaked reset | Clear all detection data |
CLI tools (available in Bash):
canary-stats # quick summary
canary-stats --json # machine-readable
canary-export --csv # export all detections
canary-export --json # export as JSON array
Add to ~/.claude/settings.json to keep the counter visible at all times:
{
"statusLine": {
"type": "command",
"command": "bash ~/.sonomos/statusline.sh"
}
}
The HUD shows your total PII count, session delta, top categories, detector breakdown, and last detection time. Color-coded by severity: green (0) / yellow (<10) / red (10+).
Drop this into your project's .claude/settings.json to auto-enable Canary for every developer:
{
"extraKnownMarketplaces": {
"sonomos": {
"source": { "source": "github", "repo": "sonomos-ai/Canary-Plugin" }
}
},
"enabledPlugins": { "canary@sonomos": true }
}
Commit it. Every team member gets prompted to install on their next session.
~/.sonomos/••0700/0600)jq to prevent injectionSee SECURITY.md for vulnerability reporting.
Most developers have no idea how much sensitive data they've shared with AI tools.
npx claudepluginhub sonomoshq/canary --plugin canaryHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Binary reverse engineering, malware analysis, firmware security, and software protection research for authorized security research, CTF competitions, and defensive security
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.