ClawHub
Back to skill

Security audit

ClankerHive

Security checks for vulnerabilities and agentic risk

Overview

ClankerHive is a disclosed local coordination store for sharing agent state, with real privacy and stale-data risks but no hidden or unrelated behavior found.

Install this only if you want local agent sessions to share state. Keep the database path private, prefer short TTLs, do not store secrets, sensitive user data, or high-trust instructions in facts or alerts, and validate any stored value before using it to trigger automated actions.

SkillSpector

Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Cross-Context Output

Medium
Category
Output Handling
Content
---
name: clankerhive
description: "Shared SQLite-backed context store for multi-session agent coordination. Use when: (1) checking if work was already done recently (email checked, briefing sent), (2) preventing duplicate cron/heartbeat runs via task claiming, (3) passing alerts between sessions (cron queues alert → main session pops it), (4) storing short-lived facts with TTL, or (5) any cross-session state sharing. Replaces ad-hoc JSON state files with a proper coordination bus. Triggers on: deduplication, cross-session state, shared facts, alert queue, task coordination, heartbeat state."
homepage: https://github.com/pfrederiksen/clankerhive
metadata:
  {
Confidence
92% confidence
Finding
This skill is explicitly designed for cross-session state sharing, which creates a real risk of unintended data exposure or trust-boundary crossing between different agent runs. Even without network access, one session can read facts or alerts written by another, so sensitive or untrusted content may persist and influence later actions in a different context.

Session Persistence

Medium
Category
Rogue Agent
Content
Instead of maintaining a separate `memory/heartbeat-state.json` file, use ClankerHive facts:

```bash
# Old way: read/write JSON file
# New way:
python3 scripts/clankerhive.py set heartbeat.email "$(date +%s)" --ttl 1800
python3 scripts/clankerhive.py set heartbeat.calendar "$(date +%s)" --ttl 3600
Confidence
91% confidence
Finding
The skill intentionally persists state across sessions in a shared SQLite database, which is a genuine persistence mechanism and can retain operational context beyond a single run. That persistence becomes dangerous if agents store secrets, user content, or untrusted instructions, because later sessions may consume stale or attacker-influenced data as if it were trusted state.

Static analysis

No suspicious patterns detected.