T05 · Unauthorized Access and Privilege Escalation
Warning
- Location
SKILL.md:132- Finding
Overbroad Access to Persistent Agent Memory and User Context
- Content
View full analysis
- Remediation
View remediation
Security audit
AI Daily Briefing
Security checks for vulnerabilities and agentic risk
Overview
This daily briefing skill is purpose-aligned, but it can surface sensitive tasks, meetings, calendar items, and agent memory from broad everyday trigger phrases without clear per-source consent.
Install only if you are comfortable with the skill reading and summarizing your task list, recent meeting notes, calendar data, and optional memory/profile files into chat output. Use explicit briefing commands, avoid invoking it while screen sharing unless the audience should see that context, and consider disabling or removing access to MEMORY.md, USER.md, or calendar sources unless you intentionally want them included.
Vulnerability Patterns
- Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
- Insecure DependenciesIntroduces malicious components through unsafe dependency sources
- 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
Findings (2)
T08 · Insecure Dependencies
Warning
- Location
README.md:7- Finding
Unpinned Remote Sources in Installation Instructions
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The skill is designed to read potentially sensitive sources including todo.md, meeting-notes/, MEMORY.md, session memory files, and calendar data, but the user-facing description says "No setup. Just say 'briefing'" without warning that personal and organizational data will be accessed. This weakens informed consent and increases the chance users invoke the skill casually, not realizing it may aggregate private tasks, meeting decisions, deadlines, and memory content into a response.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 97% confidence
- Finding
The trigger phrases include broad, natural-language requests such as "what do I need to know?", "start my day", and "give me the rundown," which can plausibly appear in unrelated conversations. In a skill-routing environment, this can cause the skill to activate unexpectedly and read from local task lists, meeting notes, memory files, or calendar sources, exposing sensitive personal or workplace context without sufficiently explicit user intent.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
Using the single word trigger "briefing" is overly generic and can easily appear in normal conversation, making accidental invocation plausible. In a skill that aggregates tasks, calendars, and meeting-note context, unintended activation could expose sensitive productivity data or cause the assistant to act in the wrong context.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The phrase "what's today look like?" is common conversational language and not clearly bound to this skill. Because this skill surfaces personal schedule and work context, broad matching raises the risk of accidental invocation and unintentional disclosure of calendar items, priorities, or recent meeting content.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The trigger "what's overdue?" is short, generic, and likely to occur in unrelated conversations about tasks or deadlines. In this skill's context, an unintended match could reveal overdue items and associated business context to the wrong user or at the wrong time.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The example activation phrases include very common conversational language such as "what's on my plate?" and "start my day," which can overlap with ordinary user speech and unintentionally trigger the skill. In a briefing skill that surfaces calendar items, overdue tasks, and meeting-derived context, accidental activation can expose sensitive personal or work information at the wrong time or in front of unintended viewers.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The quick-start text promotes automatic surfacing of to-do items, calendar meetings, conflicts, and contextual background without warning users that potentially sensitive data will be aggregated and displayed. This lack of disclosure increases the risk of inadvertent privacy exposure, especially in shared screens, public spaces, or when users do not realize linked systems are being queried automatically.
- Content
Vague Triggers
Low
- Category
- Not specified by scanner
- Confidence
- 85% confidence
- Finding
"Weekly preview" is less common than the other phrases but is still generic enough to be triggered unintentionally in routine planning discussions. The resulting disclosure risk is somewhat lower, but the skill still exposes potentially sensitive schedule and task summaries when invoked without strong boundaries.
- Content
Static analysis
No suspicious patterns detected.