Security audit

Context Management Context Save

Security checks for vulnerabilities and agentic risk

Overview

This context-saving skill is coherent, but it encourages broad project capture and persistent/vector storage without enough user control or data-safety boundaries.

Review this skill before installing if your projects contain private code, credentials, customer data, or internal architecture notes. Use it only with explicit scope, exclude secrets and sensitive files, prefer local storage unless external vector databases are approved, and define retention and deletion rules for saved context.

A.I.G

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

SkillSpector

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 (4)

Context Leakage

High

Category: Data Exfiltration
Confidence: 93% confidence
Finding: The workflow promotes periodic context snapshots, versioning, archiving, and selective restoration, which creates a clear pathway for broad context capture and retention. In this skill's context, that increases the danger because long-running archives and summaries can accumulate sensitive project state over time, making accidental leakage, over-sharing to later sessions, or unauthorized retrieval more likely.
Content: Scanner excerpt · SKILL.md (reported line 160)May include surrounding context.
md
5. Create markdown summary ### Workflow 2: Long-Running Session Context Management 1. Periodically capture context snapshots 2. Detect significant architectural changes 3. Version and archive context 4. Enable selective context restoration

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The manifest description says only 'Use when working with context management context save,' which does not define concrete triggers or boundaries beyond repeating the skill name. This ambiguity can cause unintended invocation because users and orchestrators are not given specific, narrow conditions for when the skill applies.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The listed conditions describe using the skill for its own tasks, workflows, guidance, and best practices, but they do not identify concrete user intents or domain-specific phrases. Because the guidance is circular and broad, it does not clearly distinguish when this skill should activate versus similar knowledge or planning skills.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill recommends vector database integration and semantic embedding generation for project context without clearly warning that project data may be transmitted to external services or persist in long-lived storage. In a context-management skill, this is materially risky because captured context can include source code, architecture notes, credentials, or other sensitive internal information that may be retained, indexed, or exposed beyond the immediate session.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.