Back to skill

Security audit

Feishu Calendar Advanced

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent Feishu calendar helper, but it asks users to run an unpinned external CLI that handles secrets and can delete calendar events without enough safety guidance.

Install only if you trust the @teamclaw/feishu-agent package source and are comfortable granting it Feishu calendar access. Prefer pinned package versions, avoid entering real app secrets directly in shell commands, rotate any secret already exposed in history, and verify event IDs before deletion.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (2)

T08 · Insecure Dependencies

Error
Location
SKILL.md:26
Finding

Unpinned Third-Party Package Is Downloaded and Executed

Content
View full analysis
bunx @teamclaw/feishu-agent config set appSecret ``` ```bash bunx @teamclaw/feishu-agent auth ``` ```bash bunx @teamclaw/feishu-agent whoami ``` ### Technical Analysis The Skill instructs users to execute `@teamclaw/feishu-agent` through `bunx` without specifying an exact, audited package version. In the absence of a committed lockfile or integrity verification, `bunx` may retrieve the package version currently resolved by the package registry when each command is run. Consequently, the code executed at runtime can differ from the code reviewed during this audit. A compromised package release, package-maintainer account, or registry distribution path could introduce arbitrary code. This concern is amplified because the package is used to configure a Feishu application secret, conduct OAuth authorization, and access calendar functionality. The repository contains only `SKILL.md`; it does not include the dependency's implementation or an integrity manifest. Therefore, the actual behavior of the downloaded package and the claimed conflict-detection functionality cannot be independently verified from this project. ### Attack Path 1. An attacker compromises the package maintainer's publishing credentials, package distribution channel, or a future release of `@teamclaw/feishu-agent`. 2. The attacker publishes a modified package version containing malicious installation or runtime behavior. 3. A user follows the Skill documentation and runs an unversioned `bunx @teamclaw/feishu-agent ...` command. 4. `bunx` resolves, downloads, and executes the attacker-control ...[truncated 727 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:29
Finding

Feishu Application Secret Is Passed as a Command-Line Argument

Content
View full analysis
bunx @teamclaw/feishu-agent config set appSecret ``` ### Technical Analysis The documented manual configuration procedure places the Feishu application secret directly in a command-line argument. Depending on the operating system, shell, terminal environment, and monitoring configuration, command arguments may be exposed through: - Shell history files - Process inspection utilities - Process accounting or endpoint monitoring - Terminal transcripts and session recording - Debug logs, support bundles, or copied command output The placeholder itself is not a hardcoded secret. The vulnerability arises when a user replaces it with a real secret as instructed. Command-line arguments are generally unsuitable for long-lived credentials because they may persist outside the intended credential store even after the command finishes. ### Attack Path 1. A user replaces `` with a valid Feishu application secret and executes the documented command. 2. The shell records the full command in history, or local process-monitoring software captures its arguments while it is running. 3. A local user, administrator, support operator, malware process, or party with access to collected logs retrieves the exposed argument. 4. The attacker combines the secret with the corresponding application ID. 5. The attacker authenticates as the Feishu application or otherwise abuses the credentials within the permissions assigned to that application. ### Impact Assessment Exposure may allow an attacker to impersonate the configured Feishu application within the scope of its granted permissions. Based on the documented use, affected access may include calendar and event opera ...[truncated 283 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The setup instructions tell users to store an app ID and app secret via CLI configuration without any warning that the app secret is sensitive credential material. This can lead to insecure handling, such as entering secrets in shared shells, storing them in plaintext config files, or exposing them through logs or command history.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill documents a delete operation that can remove calendar events by ID, but it provides no warning, confirmation step, or safety guidance before performing a destructive action. In an agent context, this increases the risk of accidental or unintended deletion of user data, especially if the command is invoked from ambiguous prompts or with the wrong event ID.

Content

No source excerpt is available for this finding.