Back to skill

Security audit

Sentry Integration

Security checks for vulnerabilities and agentic risk

Overview

This Sentry skill is mostly purpose-aligned, but it gives agents state-changing Sentry commands and an unverified pip install path without enough user-control safeguards.

Review before installing. Prefer the official, provenance-checked Sentry CLI distribution, avoid the pip option unless you independently trust it, use a narrowly scoped token, and require explicit confirmation before resolving, ignoring, bulk-changing issues, finalizing releases, recording deploys, or uploading source maps.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:4
Finding

Unverified Python Package Presented as an Official Sentry CLI Installation Option

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 4 and 15
Vulnerability Type: Supply-chain risk through an unverified third-party dependency
Risk Level: Medium

Vulnerable Code Snippets

Metadata declaration at line 4:

yaml
metadata: {"openclaw":{"requires":{"bins":["sentry-cli"]},"install":[{"id":"npm","kind":"npm","package":"@sentry/cli","global":true,"bins":["sentry-cli"],"label":"Install sentry-cli (npm)"},{"id":"pip","kind":"pip","package":"sentry-cli","bins":["sentry-cli"],"label":"Install sentry-cli (pip)"}]}}

Installation instruction at line 15:

bash
pip install sentry-cli

Technical Analysis

The Skill presents the unscoped PyPI package sentry-cli as an interchangeable alternative to the explicitly scoped npm package @sentry/cli. It provides no publisher verification, provenance information, exact version pin, package hash, signature validation, or other integrity control for the Python package.

Installing a package can execute package-controlled installation or build logic. The resulting sentry-cli executable is then trusted by the Skill and may run while sensitive Sentry environment variables are available. A malicious, compromised, dependency-confused, or unrelated package could therefore execute arbitrary code under the installing user's account.

This is a confirmed insecure dependency instruction. It does not establish that the current PyPI package is malicious; the vulnerability is the recommendation to install and trust it without demonstrating authoritative provenance or applying integrity protections.

Attack Path

  1. A user or agent follows the documented Python installation alternative.
  2. pip resolves the mutable package name sentry-cli from the configured Python package index.
  3. A compromised, spoofed, or otherwise untrusted release supplies package-controlled code.
  4. Code may execute during installation or when the generated sentry-cli command is invoked.
  5. The executable ...[truncated 1177 chars]
Remediation
View remediation

Remediation Suggestions

  1. Remove the pip install sentry-cli option unless the package's authoritative ownership and equivalence to Sentry's official CLI can be established.
  2. Prefer Sentry's officially documented distribution channel and link to the authoritative installation documentation.
  3. Pin the dependency to a reviewed exact version rather than resolving an unconstrained mutable version.
  4. Verify package integrity with publisher signatures, trusted provenance attestations, or cryptographic hashes where supported.
  5. Avoid global installation where possible. Use an isolated environment or a project-local, lockfile-controlled tool dependency.
  6. Configure package managers to use trusted registries and enforce dependency provenance policies.
  7. Review release notes and package ownership before upgrades, and perform periodic supply-chain scanning.
  8. Use narrowly scoped, short-lived Sentry tokens so a compromised CLI cannot obtain broader access than necessary.
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill advertises capabilities that can modify Sentry state, including resolving issues and finalizing releases, but it does not prominently warn that these operations are state-changing and may affect incident tracking or deployment records. In an agent setting, this increases the risk of unintended administrative actions being taken from a casual user request.

Content

No source excerpt is available for this finding.

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 119)May include surrounding context.

bash
# Resolve
curl -s "https://sentry.io/api/0/issues/$ISSUE_ID/" \
  -H "Authorization: Bearer $SENTRY_AUTH_TOKEN" \
  -X PUT -H "Content-Type: application/json" \
  -d '{"status": "resolved"}'

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The triage workflow instructs the agent to resolve or ignore issues once it believes fixes are deployed, but it does not require explicit confirmation or a verification gate before altering incident status. This is dangerous because an automated or over-trusting agent could suppress active incidents, degrading monitoring fidelity and hiding unresolved production problems.

Content

No source excerpt is available for this finding.