Summarize

• Outcome: Review
• Latest audit: Sep 15, 2026 · 1:51 AM UTC
• Version: 0.1.0

Overview

This summarization skill has a coherent purpose, but users should review it because it installs an unpinned third-party CLI and may send files, URLs, and media to external AI or extraction services.

Install only if you trust the Homebrew tap and the summarize CLI it provides. Do not use this skill on secrets, regulated data, private documents, or confidential URLs unless you are comfortable sending that content to the selected model provider and any enabled extraction services such as Firecrawl or Apify.

A.I.G

Vulnerability Patterns

• Insecure Dependencies: Introduces malicious components through unsafe dependency sources
• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally

Findings (1)

T08 · Insecure Dependencies

Warning

• Location: SKILL.md:5

Finding

Unverified Third-Party Homebrew Dependency

Content

View full analysis

Vulnerability Details

File Location: SKILL.md, line 5
Vulnerability Type: T08: Insecure Dependencies
Risk Level: Medium

Vulnerable Code Snippet:

yaml

metadata: {"clawdbot":{"emoji":"🧾","requires":{"bins":["summarize"]},"install":[{"id":"brew","kind":"brew","formula":"steipete/tap/summarize","bins":["summarize"],"label":"Install summarize (brew)"}]}}

Technical Analysis

The skill directs users or an agent-managed installation process to install the summarize executable from the third-party Homebrew tap steipete/tap. The formula, executable source, artifact checksum, and immutable version are not included in the audited project. Consequently, the behavior and integrity of the installed component cannot be verified from this package.

Using a mutable, externally maintained package source creates a supply-chain trust boundary. If the tap, its maintainer account, the referenced release infrastructure, or downloaded artifacts are compromised, installation may deliver code that differs from the code originally reviewed. The project does not pin an immutable formula revision or independently verify an expected artifact digest.

This finding does not establish that the current third-party formula is malicious. It identifies that the skill delegates executable installation to an unreviewed external source without integrity controls visible in the audited files.

Attack Path

1. An attacker compromises the third-party Homebrew tap, its maintainer credentials, or an artifact distribution endpoint used by the formula.
2. The attacker modifies the formula or a referenced binary/archive to include malicious installation or runtime behavior.
3. A user or agent loads the skill and follows its declared installation metadata to obtain the missing summarize executable.
4. Homebrew resolves and executes the third-party formula and downloads the attacker-controlled component.
5. The malicious installati ...[truncated 780 chars]

Remediation

View remediation

Remediation Suggestions

1. Pin the dependency to an immutable, reviewed release rather than a mutable tap reference.
2. Verify downloaded artifacts against a cryptographic SHA-256 or stronger digest maintained in a trusted location.
3. Vendor the formula and relevant executable source into a reviewable repository, or use a trusted package registry with provenance and signing support.
4. Require signed releases and verify signatures before installation.
5. Run installation and execution with least privilege; do not use an administrator account unless strictly necessary.
6. Restrict the CLI process to only the files, environment variables, and network destinations required for summarization.
7. Add automated dependency provenance checks, checksum validation, and periodic review of the upstream formula and release artifacts.
8. Clearly disclose that URLs, local files, and potentially sensitive content may be processed by external model or extraction providers before users invoke the tool.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 98% confidence

Finding

The skill encourages summarizing URLs, local files, PDFs, images, audio, and YouTube content using third-party model providers and optional extraction services, but it does not clearly warn users that submitted content may be transmitted off-device to external APIs. This creates a real privacy and data-handling risk because users may provide sensitive local files or confidential URLs without understanding that their contents could be sent to OpenAI, Anthropic, Google, xAI, Firecrawl, or Apify.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.