Security audit

QMD - Quick Markdown Search

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent local Markdown search helper, with a disclosed but avoidable supply-chain risk in its install command.

Install only if you are comfortable trusting the qmd upstream repository and Bun global install path. Prefer pinning the install to a reviewed commit or release, avoid sudo, and index only Markdown folders you intentionally want searchable.

A.I.G

Vulnerability Patterns

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
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (1)

T08 · Insecure Dependencies

Warning

Location: SKILL.md:5
Finding: Unpinned Remote Dependency Installed Globally from a Mutable Git Repository
Content: View full analysis
Remediation: View remediation
``` 2. Prefer a signed, versioned release from a trusted package registry when available. Pin the exact version rather than using a range or moving tag. 3. Verify downloaded artifacts using a published cryptographic checksum or signature before installation. 4. Record the reviewed version, commit hash, expected checksum, and upstream source in `SKILL.md` so installations are reproducible and auditable. 5. Use package-manager provenance or signature verification where supported, and reject installation when verification fails. 6. Avoid elevated privileges. Perform installation under a dedicated, least-privileged user account or inside a sandbox/container when practical. 7. Review dependency installation hooks and the resulting executable before recommending an updated version or commit. ]]>

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Static analysis

No suspicious patterns detected.