Security audit

recurring-content-calendar

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly disclosed PostNitro workflow for drafting, auditing, and scheduling recurring social posts, with explicit user approval before live scheduling.

Install this only if you want Codex to help operate a PostNitro-based social publishing workflow. Review the selected accounts, draft status, recurring cadence, and scheduled/public approval steps carefully because it can affect live brand channels after you approve scheduling.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Context-Inappropriate Capability

Medium

Confidence: 80% confidence
Finding: The instruction to use jq or a subagent on overflowing post-history data expands processing beyond the core social-scheduling workflow and can expose large volumes of workspace data to additional tooling or model contexts. In this skill, the history feed may include captions, account metadata, statuses, and error details across many posts, so delegating or externally transforming that data increases data-minimization and unintended disclosure risk.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The invocation description is broad enough to trigger on generic social-media planning requests, causing the agent to enter a workflow that can connect PostNitro, inspect workspace assets, and prepare scheduling actions even when the user only wanted advice or draft content. Because this skill ultimately supports recurring automated publishing to live accounts, overbroad matching raises the risk of unintended tool use and excessive access to connected account data.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The skill begins by describing an always-running content calendar without an upfront, prominent warning that it may schedule posts to real social accounts and create recurring automation. In this context, insufficient notice is dangerous because users may engage what seems like a content-generation skill but actually authorize actions affecting public brand channels and persistent automated behavior.

Static analysis

No suspicious patterns detected.