Versioned security assessment

Report ID: SA-305A75C8

6/27/2026, 5:19:56 PM

mac-automation security assessment v4

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
mac-automation
Version
v4
Maintainer
7Sageer, Claude
Source
github.com/7Sageer/cc-assistant/tree/main/.claude/skills/mac-automation
Coverage
9 Files scanned · 2,301 Lines analyzed
Policy version
Unavailable

Highest confirmed finding severity

Medium

1 confirmed security finding requires attention.

Installation context

Historical evidence

This report may not describe the currently installable artifact. Open the current Skill page for install guidance.

Open current Skill page

This report does not block or authorize the manifest or ZIP.

Static analysis correctly identified extensive external command usage, but this is the declared purpose of the skill: guiding osascript and AppleScript automation. I found no evidence of prompt injection, hidden network exfiltration, malware staging, or confirmed malicious intent. The skill should publish with a warning because it can read private local data, manipulate files, send mail, run shell commands through AppleScript, and perform disruptive system actions.

Report position

Historical report

Open audit history before using this report to install.

Audit attestation

Not attestable

The required immutable binding is incomplete.

Human verification

Not verified

No human verification is recorded for this report.

Coverage

9 Files scanned · 2,301 Lines analyzed

5 items shown for review

Limitations

This report does not claim runtime or sandbox execution and does not prove the absence of side effects.

Evidence chain

Follow the evidence from source binding to the install contract. Available evidence supports verification; it is not a safety guarantee.

  1. Source

    Binding unavailable

  2. Artifact

    Identity incomplete

  3. Audit

    Complete

  4. Install contract

    Open manifest to verify

    Open manifest

Capabilities observed

Observed means this report recorded supporting evidence. Not recorded does not prove that a capability is absent.

Contains scripts

May execute code included with the Skill.

Not recorded by this audit

Network access

May connect to external services.

Observed in 4 evidence locations

Filesystem access

May read or write local files.

Observed in 3 evidence locations

Env variables

May read values from the process environment.

Not recorded by this audit

External commands

May invoke commands or programs outside the Skill.

Observed in 3 evidence locations

Capability review items (4)
Medium
SKILL.md:10-22references/system-applescript.md:219-253references/safari-applescript.md:299-303
Broad Local Command Execution Through AppleScript
TRUE_POSITIVE: The skill teaches agents to run osascript from Bash and includes AppleScript do shell script examples. This is legitimate for a macOS automation skill, but it can execute local commands and inherits the user permissions granted to the agent and macOS Automation.
The command execution pattern is explicit and central to the skill. The surrounding context shows intended local automation rather than hidden execution, so the risk is real but dual-use.
Medium
references/finder-applescript.md:98-118references/mail-applescript.md:163-171references/system-applescript.md:263-280
Destructive and Disruptive Automation Examples
TRUE_POSITIVE: The skill includes examples for deleting Mail, reminders, calendar events, calendar lists, files, emptying Trash, and power actions such as restart or shutdown. These are expected automation capabilities but require confirmation and clear user intent.
The examples perform concrete destructive or disruptive actions. The skill documents confirmation for destructive operations, so this is a controlled dual-use risk rather than confirmed abuse.
Medium
references/system-applescript.md:229-236references/system-applescript.md:256-260references/system-applescript.md:283-296
Screen Capture and Keyboard Simulation Capabilities
NEEDS_REVIEW: Static keylogger and screen capture upload alerts are overstated because the files show screenshot creation and keystroke simulation, not key capture or upload. These capabilities are still privacy-sensitive and can affect active applications.
The cited lines contain screenshot and keystroke commands, but no evidence of uploading screenshots or recording keystrokes. The remaining risk is legitimate automation that can still be misused.
Medium
references/safari-applescript.md:152-176references/safari-applescript.md:296-304
Safari JavaScript Execution and Form Interaction
TRUE_POSITIVE: Safari references execute JavaScript in the active tab, click page elements, and fill form fields. This can automate useful browser tasks, but it can also alter web pages or interact with authenticated sessions if misused.
The JavaScript execution and form manipulation examples are explicit. No malicious target or credential exfiltration is shown, so this is a medium dual-use browser automation concern.

Risk findings

Confirmed security concerns are separated from items that still need review.

Confirmed security concerns (1)

RISK-001 Medium
references/mail-applescript.md:64-82SKILL.md:124-129references/safari-applescript.md:64-75
Access to Private Local Data
TRUE_POSITIVE: The references include reading unread Mail metadata, Calendar data, clipboard contents, Safari page source, and selected page text. This can expose sensitive personal or business information if used without narrow user consent.
The cited examples directly retrieve local application data and browser content. The skill also instructs privacy-sensitive data to be displayed only when requested, which reduces malicious intent but not exposure risk.

Expert evidence

Immutable subject identity, scanner metadata, dismissed matches, and source-level evidence.

Artifact subject

Marketplace commit
Unavailable
Content hash
Unavailable
Tree hash
Unavailable
Skill path
Unavailable
Audit payload hash
Unavailable

Analysis metadata

Audit model: codex

Analysis state: Complete

Scope is limited to the recorded files, lines, methods, and evidence. No runtime or sandbox execution is claimed.

Static false positives ignored (3)
Low
references/safari-applescript.md:7-13references/safari-applescript.md:19-33references/safari-applescript.md:39-43
Hardcoded URL Alerts Are Benign Examples
FALSE_POSITIVE: The Safari URLs point to common example destinations such as Google, Apple, GitHub, and example.com. I found no evidence that these URLs receive local files, secrets, screenshots, or private application data.
The URLs are visible sample navigation targets and search examples. No upload, webhook, credential, or suspicious domain evidence was found in the reviewed context.
Low
SKILL.md:1-5references/calendar-applescript.md:20-30references/system-applescript.md:391-395
Weak Cryptography Alerts Are False Positives
FALSE_POSITIVE: The reported weak cryptography locations are ordinary AppleScript or documentation text, not MD5, SHA1, or cryptographic code. These alerts appear to be pattern matches against unrelated words or Markdown structure.
The cited locations contain frontmatter, event creation examples, and tips, with no cryptographic algorithm usage. This strongly supports a false-positive classification.
Low
SKILL.md:80-86
No Prompt Injection Attempt Found
FALSE_POSITIVE_CHECK: I checked for text that tries to override evaluator instructions, claim pre-approval, or skip analysis. No evidence found in the reviewed skill files.
Targeted searches did not find injection phrases. Confidence is high enough for this audit, but it is based on reviewed files and targeted checks rather than exhaustive semantic proof.

Verify and export

JSON report CITATION.cff SARIF 2.1.0 Signed installation manifest

The manifest and lockfile bind install artifacts to cryptographic hashes. This integrity claim is separate from the security assessment.

Audit attestation: not_attestable