Back to skill

Security audit

Dataecho Memory

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed cloud-backed agent memory tool, but users should understand that saved memories persist outside the session on DataEcho.

Install this only if you want agent memories stored in DataEcho cloud storage. Do not save secrets, API keys, regulated personal data, or confidential business information, and prefer explicit confirmation before saving new memories or issuing writable handoff tokens.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs the agent to trigger on broad natural-language cues such as 'remember this', 'don't forget', and 'like last time'. That can cause unintended activation and silent persistence of user content to an external cloud memory service without a sufficiently explicit, informed action from the user.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The description emphasizes durable memory in a private cloud drive but does not prominently warn that user data may be transmitted to and stored on a third-party service outside the local session. Users may reasonably interpret 'memory' as internal agent state rather than external cloud persistence, creating a consent and privacy risk.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
93% confidence
Finding

The skill is explicitly built for persistent cross-session storage of user and project information, including handoff to future agents and other machines. That persistence increases privacy and data-governance risk because retained content may outlive user expectations, be recalled in later contexts, or be exposed through sharing features and account compromise.

Content

Scanner excerpt · SKILL.md (reported line 67)May include surrounding context.

md
One fact per file. Slug is lowercase-kebab and *says what the fact is*
(`user-prefers-tabs`, not `note-1`). Description ≤ 100 chars — it is the recall surface,
write it like a search result. Body under ~8 KB (the script warns; split bigger things).
For decisions and feedback, include **Why:** and **How to apply:** lines — a decision
without its why gets re-litigated next session. Link related memories with `[[slug]]`.

External Transmission

Medium
Category
Data Exfiltration
Confidence
87% confidence
Finding

The skill includes commands that transmit identifiers and authentication data to an external service and is fundamentally designed to store session memory remotely. External transmission is the core behavior here, so the danger is not the curl syntax itself but that potentially sensitive user context could be uploaded off-platform if the agent follows the skill automatically.

Content

Scanner excerpt · SKILL.md (reported line 81)May include surrounding context.

$ARTIFACT_API_KEY exists, you're done. Otherwise (same flow as the dataecho skill):

bash
curl -sS -X POST https://dataecho.ai/api/auth/agent/request-code -H 'content-type: application/json' -d '{"email":"user@example.com"}'
# user reads the emailed code, then:
curl -sS -X POST https://dataecho.ai/api/auth/agent/verify-code -H 'content-type: application/json' -d '{"email":"user@example.com","code":"<CODE>"}'
echo '<API_KEY>' > ~/.artifact/credentials && chmod 600 ~/.artifact/credentials

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · SKILL.md (reported line 84)May include surrounding context.

curl -sS -X POST https://dataecho.ai/api/auth/agent/request-code -H 'content-type: application/json' -d '{"email":"user@example.com"}'

user reads the emailed code, then:

curl -sS -X POST https://dataecho.ai/api/auth/agent/verify-code -H 'content-type: application/json' -d '{"email":"user@example.com","code":""}' echo '<API_KEY>' > ~/.artifact/credentials && chmod 600 ~/.artifact/credentials

text

The `apiKey` is returned **once** — write it to the file immediately; don't echo it.