Back to skill

Security audit

Dogearai Memory

Security checks for vulnerabilities and agentic risk

Overview

The skill does what it claims, but it automatically creates and persists a remote memory account, sends user context to DogearAI under broad triggers, and exposes tokens in ways users should review carefully.

Review DogearAI's data handling before installing. Use this only if you are comfortable with remote memory storage, local token persistence, and agent-triggered recall/write behavior. Do not store secrets, customer data, regulated data, or confidential project details unless you have explicitly accepted that risk, and avoid running it in environments where DOGEAR_BASE_URL or command output could be controlled or logged by others.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (3)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:58
Finding

Automatic External Transmission of User Information Without Explicit Per-Write Consent

Content
View full analysis
"`. **Save:** preferences ("I prefer X", "always Y"), decisions ("we're going with X", "we ruled out Y"), facts about the user / project / stack / goals, commitments and TODOs. **Don't save:** transient chatter, anything trivially re-derivable from the current files/code, or secrets — unless the user asks. The server classifies the memory and files it into the right space for you — **you don't pick the space**. One call = one atomic memory; split long, multi-topic notes into several calls. A short "Saved to DogearAI" is enough. ## Etiquette - Reads (`context`, `spaces`, `read-space`) are safe and silent — no need to ask. ``` `dogear.py:138-144`: ```python elif a.cmd == "remember": body = {"text": a.text} if a.source: body["source"] = a.source if a.raw: body["mode"] = "raw" st, out = call("POST", "/v1/memories", body) ``` ### Technical Analysis The Skill instructs the Agent to identify information it considers durable and automatically submit the user's words to DogearAI. The implementation serializes the supplied text and transmits it to the external `/v1/memories` API. Persistent cloud memory is part of the declared functionality, so network transmission itself is expected. However, the instruction does not require explicit confirmation before each write, does not show the user the exact content that will be uploaded, and delegates sensitivity classification to the Agent. The exclusion for secrets does not cover confidential project details, personal information, internal decisions, customer informa ...[truncated 1648 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
dogear.py:22
Finding

Bearer Token and Memory Data Can Be Redirected to an Environment-Controlled Origin

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
dogear.py:74
Finding

API Bearer Token Is Printed in Claim URLs

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (10)

Tainted flow: 'req' from os.environ.get (line 32, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · dogear.py (reported line 38)May include surrounding context.

python
if data is not None:
        req.add_header("Content-Type", "application/json")
    try:
        with urllib.request.urlopen(req, timeout=30) as r:
            return r.status, r.read().decode("utf-8")
    except urllib.error.HTTPError as e:
        return e.code, e.read().decode("utf-8")

Missing User Warnings

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill promotes 'zero setup' external memory syncing and automatic account/token creation without giving a prominent upfront warning that user context may be transmitted to and stored by a third-party service. In this skill's context, that omission is especially risky because the stated purpose is to persist long-term personal and project information across tools, which may include sensitive preferences, decisions, TODOs, and other durable context.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The README explicitly says the skill will recall and save user context and durable facts via the DogearAI API, but it does not clearly and prominently warn that potentially sensitive project details, personal preferences, decisions, and other durable context are transmitted to a third-party remote service. In a memory skill, users may assume data stays local or within the current tool, so the missing disclosure increases the risk of inadvertent sensitive-data exfiltration and privacy harm.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill instructs the agent to execute a local Python script that can access environment variables, write files, use the network, and invoke shell commands, yet the manifest declares no explicit tool restrictions. That mismatch increases the chance the skill will run with broader privileges than users expect, including auto-creating accounts, persisting tokens locally, and transmitting user data to an external service.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill is described as usable 'at the start of any task' needing saved context and whenever the user says something durable, which is broad enough to trigger in many normal conversations. That expansive activation guidance can cause unnecessary external recalls and writes of user data, increasing privacy risk and the chance of silent background data transfer.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The recall guidance treats common phrases like 'my project', 'my preferences', 'the stack', or 'as we decided' as triggers to query the external memory service before clarifying intent. In context, this is more dangerous because the skill explicitly says reads are 'safe and silent,' encouraging covert retrieval of potentially sensitive long-term context based on ordinary language.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The CLI writes authentication tokens to ~/.dogear/token, including tokens obtained via automatic registration, without a user-facing confirmation before persistence occurs. Although permissions are tightened best-effort, silent credential persistence can violate user expectations and increases the chance of unintended long-term session retention on shared or managed systems.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

When no token is present, the tool silently creates a remote DogearAI account and persists its token locally as a side effect of ordinary commands. In a memory skill, this is more dangerous because users may assume local-only recall/persistence, while the code causes remote account creation and data association without explicit prior consent.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
91% confidence
Finding

The code establishes persistent session state by auto-registering an account and saving its bearer token for reuse. Persistent authentication is not inherently unsafe, but doing it automatically in a cross-tool memory skill increases privacy and account-retention risk because future invocations may continue linking user data to the same remote identity without fresh consent.

Content

Scanner excerpt · dogear.py (reported line 60)May include surrounding context.

python
def _auto_register():
    """No token anywhere -> create a zero-config anonymous account, save & return its token."""
    status, out = _http("POST", "/v1/auth/anonymous", body={"label": "auto"})
    if status >= 400:
        sys.exit(f"ERROR: could not auto-create a DogearAI account ({status}).\n{out}")

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The remember/context command flow sends user memory content to a remote service without an explicit warning at the time of use. Given this skill’s purpose of storing durable personal or project context, the transmitted data may be sensitive, making undisclosed remote exfiltration of prompts, preferences, or project details materially risky.

Content

No source excerpt is available for this finding.