Back to skill

Security audit

Obsidian Karpathy Wiki

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Obsidian vault maintenance skill, but users should understand that it is designed to reorganize and write local notes when used.

Install this only if you want an agent to actively maintain an Obsidian vault. Before broad runs, specify the vault path, whether loose root notes may be moved, what folders are in scope, and whether the agent should ask before saving outputs, updating formal pages, or archiving duplicates.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (6)

Direct Prompt Extraction

High
Category
System Prompt Leakage
Confidence
85% confidence
Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · references/heartbeat.md (reported line 39)May include surrounding context.

md
- identify stale governance and stale assumptions
- review whether the object model still fits the domain

## Output Rule

Heartbeat should produce one of two outcomes:

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The description says to use the skill whenever an agent needs to ingest raw notes, reconcile inbox material, maintain summaries, compile pages, save outputs, run maintenance, lint contradictions, or hand off wiki operations. This is a very wide set of generic activities without explicit trigger boundaries or exclusion conditions, increasing the chance of unintended invocation in normal note-management contexts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly instructs the agent to move loose notes, create or update source pages, modify formal pages, and append logs, but it does not require explicit confirmation before altering user data. In a real agent environment, this can lead to unintended file modification, content relocation, or destructive reorganization of a user's vault when the user may have expected analysis only.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The default prompt is broad enough to match many ordinary note-taking, wiki, or knowledge-base tasks without clearly limiting when the skill should be invoked. In an agentic environment, this can cause over-selection of a skill that performs persistent vault operations, increasing the chance of unintended file modifications or use in contexts the user did not explicitly authorize.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill advertises writing back to the Obsidian knowledge base but does not warn that it may modify local files or require explicit user approval before doing so. This creates a real risk of silent or surprising persistence, especially because the skill is designed around maintaining a long-lived vault and may write synthesized or inferred content into user-managed notes.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The phrase "Use this when new material enters the vault" is an ambiguous invocation condition for a markdown skill document. It does not define what counts as "new material," when the workflow should not be used, or any negative examples, which could cause overly broad or unintended use.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.