Back to skill

Security audit

Work Progress Summary Pro

Security checks for vulnerabilities and agentic risk

Overview

This skill transparently manages a local SQLite work log, including edits and deletions, without evidence of hidden access, exfiltration, or unrelated behavior.

Install this if you want Codex to keep a local work-log database. Before letting it replace a day, clear a day, or delete an entry, confirm the exact date or entry id because those actions change the current report, even though history is retained.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (6)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding

The skill instructs use of a runtime script and references environment-variable behavior, but it declares no explicit tool scope or allowed-tools boundary. That omission increases the chance an agent will invoke broader execution capabilities than necessary, reducing least-privilege protections and making abuse or accidental misuse harder to constrain.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description advertises replace, edit, and delete operations but gives no explicit requirement to confirm destructive actions with the user before execution. In an agent context, ambiguous user phrasing or prompt injection in surrounding conversation could lead to unintended loss or modification of local work-history data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The workflow operationalizes overwrite and delete commands directly, and even instructs the agent to stop only on command failure, without requiring a safety checkpoint for data-loss actions. Because the skill manages persistent local records, a mistaken or manipulated invocation could erase or replace legitimate historical entries while still appearing to follow the documented workflow.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The Chinese trigger examples are very broad, everyday phrases about completed work and summaries, which increases the chance the skill activates on ordinary conversation without clear user intent to write to or modify a persistent local database. Because this skill can record, replace, update, and delete stored work logs, overbroad activation can lead to unintended data creation or modification from ambiguous prompts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The reference includes destructive operations like replace-day and delete-entry but does not instruct the agent to warn the user that stored work-log data will be overwritten or removed. In a skill backed by a local SQLite database with history tracking, missing pre-action warnings or confirmations can cause accidental loss or unintended mutation of user records when prompts are vague or referential like 'delete that one'.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This markdown file explicitly instructs users to clear a day's report by replacing it with an empty list, which can remove previously recorded data. The description gives the command shape but does not include any warning that this action is destructive or may permanently overwrite prior entries.

Content

No source excerpt is available for this finding.