ClawHub
Back to skill

Security audit

Slack Messaging

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward Slack API helper skill, but users should be careful because it can post to Slack and read workspace data when given broad token scopes.

Install only if you are comfortable giving an agent a Slack token. Prefer a bot token with the minimum scopes needed for your workflow, avoid user tokens unless required, do not use it on confidential channels without approval, and review any file links or message content before posting.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (7)

Context Leakage

High
Category
Data Exfiltration
Confidence
87% confidence
Finding

The workflow encourages collecting threaded replies, which may aggregate employee responses and operational details without any data-minimization or retention guidance. In an agent-assisted environment, this can cause unnecessary collection of internal status updates or personal information and increase context leakage into automated processing pipelines.

Content

Scanner excerpt · SKILL.md (reported line 143)May include surrounding context.

md
**Incident channel:** Create a private channel named `incident-YYYY-MM-DD-<slug>`, set the topic, invite responders, post the initial report with severity and affected services.

**Standup reminder:** Schedule a daily message at 9:30 AM to #team with a prompt template. Collect threaded replies.

**Error alert:** When Sentry fires a webhook, post the error title, count, and link to the relevant channel. Add a 🔴 reaction for critical, 🟡 for warning.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill enables transmission and retrieval of workspace messages, channel metadata, user directory data, and files without warning about privacy, authorization, or data-handling constraints. In an agent setting, this omission can normalize use of sensitive Slack scopes and cause users to expose internal communications or personal data without informed consent.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 14)May include surrounding context.

Setup

bash
# 1. Create a Slack App: https://api.slack.com/apps → Create New App
# 2. Add scopes under OAuth & Permissions:
#    Bot Token Scopes: chat:write, channels:read, channels:history,
#                      files:write, reactions:write, users:read

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 29)May include surrounding context.

bash
# Simple text message
curl -s -X POST https://slack.com/api/chat.postMessage \
  -H "Authorization: Bearer $SLACK_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"channel": "C0123ABCDEF", "text": "Deploy complete ✓"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 85)May include surrounding context.

md
-H "Authorization: Bearer $SLACK_TOKEN" | jq '.channels[] | select(.name=="general") | .id'

# Create a channel
curl -s -X POST https://slack.com/api/conversations.create \
  -H "Authorization: Bearer $SLACK_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "incident-2026-04", "is_private": false}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
88% confidence
Finding

Adding a remote file link can cause Slack users to access external content and can be abused to distribute untrusted links or route workspace users to third-party systems. In an agent context, this creates a data-governance and phishing-adjacent risk because external URLs may be inserted into trusted collaboration channels without validation.

Content

Scanner excerpt · SKILL.md (reported line 109)May include surrounding context.

md
# Then POST the file content to upload_url, then call files.completeUploadExternal

# Share a remote file link
curl -s -X POST https://slack.com/api/files.remote.add \
  -H "Authorization: Bearer $SLACK_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"external_id": "report-1", "external_url": "https://example.com/report.pdf", "title": "Monthly Report"}'

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The manifest emphasizes Slack messaging, channel management, file uploads, reactions, and notifications, which are primarily write/automation workflows. The file also documents reading channel history and performing user-directory lookups, including lookup by email, which expands into workspace reconnaissance and data retrieval not clearly conveyed by the high-level description.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.