T09 · Insecure Skill Coding Practices
Error
- Location
SKILL.md:87- Finding
Path Traversal Through Unsanitized Company Slug
- Content
View full analysis
- Remediation
View remediation
Security audit
Interview Prep
Security checks for vulnerabilities and agentic risk
Overview
The skill has a coherent interview-prep purpose, but it needs review because it reads private email/tracker data and writes sensitive interview codes and contacts to disk with under-scoped safeguards.
Install only if you are comfortable letting the skill read interview-related Gmail, Notion, tracker, profile, and local note data. Before using it, confirm the exact output path, avoid untrusted company-name input, and do not let it save access codes, meeting links, or personal contact emails unless you explicitly want that information stored in plaintext.
Vulnerability Patterns
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- 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
Findings (3)
T09 · Insecure Skill Coding Practices
Error
- Location
SKILL.md:129- Finding
SQL Injection Through Interpolated Company Name
- Content
View full analysis
- Remediation
View remediation
T09 · Insecure Skill Coding Practices
Warning
- Location
SKILL.md:94- Finding
Sensitive Interview Credentials Persisted in Plaintext
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Rogue AgentSelf-Modification, Session Persistence
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
The skill description advertises cross-system data access to Gmail, Notion, and local notes, but provides no explicit privacy warning, consent boundary, or minimization guidance before collecting and consolidating interview-related personal data. That increases the risk of users triggering the skill without understanding that private emails, contact details, and scheduling information will be accessed and copied into a local document.
- Content
Session Persistence
Medium
- Category
- Rogue Agent
- Confidence
- 60% confidence
- Finding
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
- Content
md Trigger when the user asks to: - Prepare for an upcoming interview ("prep notes for [Company] interview", "create interview prep for [Company]") - Generate a prep document for a scheduled interview - Update or add to an existing prep doc for a company - Research a company before an interview
Ssd 3
Medium
- Category
- Not specified by scanner
- Confidence
- 98% confidence
- Finding
The skill directs the agent to fetch Gmail threads and extract sensitive details such as access codes, contact identities, interview times, and ATS links, then merge them into a consolidated prep document by default. Centralizing these secrets and personal details in a local markdown file materially increases exposure risk through local compromise, accidental sharing, indexing, backups, or later model access.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The workflow instructs the agent to create directories and write or update files in a fixed local path, including merge behavior, without an explicit warning or confirmation step for filesystem modification. This can lead to unintended persistence of sensitive interview data, accidental overwrites, or changes to existing notes the user did not expect.
- Content
Ssd 3
Medium
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The output template explicitly requires inclusion of access codes, contact emails, and day-of logistics pulled from emails, which operationalizes persistence of sensitive secrets and personal data in the final artifact. Because the file is stored in a predictable local project directory, the context makes this more dangerous by creating a durable, discoverable repository of interview credentials and contact information.
- Content
Static analysis
No suspicious patterns detected.