Back to skill

Security audit

skill-a-create-meeting

Security checks for vulnerabilities and agentic risk

Overview

This meeting skill largely matches its stated purpose, but it needs review because it handles Gitea tokens and repository data with unsafe setup and transport practices.

Review before installing. Use only an HTTPS Gitea endpoint, rotate any token used with the provided HTTP default, scope the bot token to the minimum repositories and actions, avoid running setup as root, and treat generated invitation content as untrusted until HTML escaping and URL validation are added.

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 (4)

T09 · Insecure Skill Coding Practices

Error
Location
env-example.txt:2
Finding

Gitea access token transmitted over plaintext HTTP

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
setup.sh:28
Finding

Dotenv configuration file is executed as arbitrary shell code

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/email_utils.py:19
Finding

Unescaped meeting data permits HTML injection into invitation emails

Content
View full analysis
会议主题 {topic} ``` ```python 🎥 加入腾讯会议 📝 查看 / 编辑议程 ``` Other interpolated values in the same template, including `meeting_code`, `organizer`, and `repo`, are also inserted without escaping. ### Technical Analysis The email generator directly interpolates data into HTML text nodes and quoted attributes. No HTML escaping is applied, and URL schemes or hosts are not validated. An attacker-controlled topic can inject arbitrary markup into the message body. A crafted URL containing a quote can terminate the `href` attribute and introduce additional attributes or elements. The resulting HTML is returned to the orchestration layer and is intended to be sent to all resolved repository-member email addresses. Email clients often restrict scripts, but injected links, images, formatting, tracking resources, and deceptive content remain practical. The vulnerability can therefore facilitate phishing and content spoofing even when JavaScript is blocked. ### Attack Path 1. An attacker supplies or influences a meeting topic, organizer value, repository value, meeting code, or join URL. 2. The value contains HTML or attribute-breaking syntax, such as: ```text R ...[truncated 956 chars]
Remediation
View remediation

T08 · Insecure Dependencies

Warning
Location
requirements.txt:1
Finding

Unpinned dependencies are installed into the system Python environment

Content
View full analysis
=2.28.0 python-dotenv>=1.0.0 PyYAML>=6.0 pytz>=2023.3 python-dateutil>=2.8.2 ``` ```bash pip install -r "$SKILL_DIR/requirements.txt" --break-system-packages -q ``` ### Technical Analysis All dependencies use open-ended minimum-version constraints. A future version can therefore be installed without having been reviewed with this Skill. The installation also uses `--break-system-packages`, bypassing protections for the operating system’s managed Python environment. This creates a non-reproducible supply-chain boundary: the code executed at installation and runtime can change even though the Skill package itself has not changed. Installing into the system interpreter can also replace or conflict with packages used by unrelated applications. No malicious dependency or typosquatted package was identified in the current list. The risk arises from unsafe version and installation practices. ### Attack Path 1. A user runs `bash setup.sh`. 2. `pip` resolves any package version satisfying the open-ended constraints. 3. A compromised, malicious, or incompatible future release is selected from the configured package index. 4. Package build or installation logic executes during installation, or malicious behavior executes when the package is imported. 5. Because the installation is not isolated, the package can affect this Skill and other Python applications using the same interpreter. A separate availability path exists where dependency upgrades overwrite system-managed versions and break unrelated software. ### Impact Assessment The execution privileges are those of the user running setup. If setup is run with elevated privileges, package installation code may receive system-level access. Potential effects include: - Executio ...[truncated 350 chars]
Remediation
View remediation
Vulnerability Patterns
  • 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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (30)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

A skill described as a full meeting orchestration entry point but lacking the promised controls can cause the surrounding agent to make unsafe assumptions, such as believing Tencent meeting creation, member lookup, and email preparation were completed or validated when they were not. Security-sensitive workflows become fragile when documentation substitutes for actual enforcement and implementation.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

A skill described as a full meeting orchestration entry point but lacking the promised controls can cause the surrounding agent to make unsafe assumptions, such as believing Tencent meeting creation, member lookup, and email preparation were completed or validated when they were not. Security-sensitive workflows become fragile when documentation substitutes for actual enforcement and implementation.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

A skill described as a full meeting orchestration entry point but lacking the promised controls can cause the surrounding agent to make unsafe assumptions, such as believing Tencent meeting creation, member lookup, and email preparation were completed or validated when they were not. Security-sensitive workflows become fragile when documentation substitutes for actual enforcement and implementation.

Content

No source excerpt is available for this finding.

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/create_meeting.py (reported line 30)May include surrounding context.

python
SKILL_DIR="$(cd "$(dirname "$0")" && pwd)"
CONFIG_DIR="$HOME/.config/skill-a-create-meeting"
ENV_FILE="$CONFIG_DIR/.env"

echo "🚀 设置 Skill-A (create_meeting)..."
echo ""

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/create_meeting.py (reported line 176)May include surrounding context.

python
SKILL_DIR="$(cd "$(dirname "$0")" && pwd)"
CONFIG_DIR="$HOME/.config/skill-a-create-meeting"
ENV_FILE="$CONFIG_DIR/.env"

echo "🚀 设置 Skill-A (create_meeting)..."
echo ""

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · setup.sh (reported line 6)May include surrounding context.

sh
SKILL_DIR="$(cd "$(dirname "$0")" && pwd)"
CONFIG_DIR="$HOME/.config/skill-a-create-meeting"
ENV_FILE="$CONFIG_DIR/.env"

echo "🚀 设置 Skill-A (create_meeting)..."
echo ""

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill advertises shell/network/env-backed behavior but does not declare any explicit tool scope such as allowed-tools or permissions. In an agent runtime, this can lead to over-privileged execution where the skill gains broader access than users or reviewers expect, increasing the blast radius for misuse of environment secrets, shell commands, or network operations.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger criteria are extremely broad and map to ordinary conversational phrases like 'open a meeting' or 'arrange a meeting,' making over-triggering likely. In an agent system, broad activation can cause unintended repository writes, member/email lookups, and downstream meeting/email actions from ambiguous user requests that did not clearly consent to these side effects.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The examples encourage activation across many common phrasings without corresponding negative examples or constraints, which increases the chance that the agent will invoke the skill when the user intended only discussion, brainstorming, or a query. Because this skill coordinates downstream state changes and email preparation, accidental invocation can leak participant data or create unwanted artifacts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill description does not clearly warn users that invoking it may create repository files, write logs, and process member email addresses for outbound invitations. This weakens informed consent and transparency around data handling, especially because the workflow touches collaboration metadata and potentially personal contact information.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The file comments specify the workflow responsibilities entirely in Chinese, indicating a language-specific operating assumption for the skill. Under the policy rule, forcing or presuming a specific language without offering user choice or documenting a justified locale constraint is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The script hard-codes the timezone to Asia/Shanghai and uses Chinese-only CLI descriptions and output messages throughout the file. This is a natural-language locale policy concern because the skill forces a specific language/locale experience without offering the user any choice or opt-in.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
78% confidence
Finding

The script creates meta.yaml and agenda.md in the target repository, which is a file-writing operation affecting user/project data. Although the module docstring describes that it manages meeting files, there is no explicit user-facing warning or confirmation near execution that repository contents will be created or modified for the specified repo.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The generated email hard-codes lang="zh" and uses Chinese-only user-facing text such as the heading and status message. This imposes a specific language/locale in the skill output without any visible opt-in or documented region-specific justification, which matches the language/locale policy violation criteria.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

gitea_request sends an authorization token in HTTP headers and is used for API calls involving repository contents and user email lookups, but the function has no docstring, comment, or user-facing logging that warns about remote transmission of sensitive data. This matches the missing-warning criterion for code files because network calls carrying credentials or user/system data should have some disclosure when not otherwise documented in the file.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The functions create_file_in_repo and update_file_in_repo perform remote file writes to a Gitea repository, which can change user or project data, but the code provides no confirmation prompt, logging, or warning comment/docstring describing this effect. For code files, safety-relevant write operations should have some visible disclosure unless clearly warned elsewhere; no such disclosure is present in this file.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This code performs outbound HTTP requests to fetch and update a remote log file, sending both log contents and an authorization token. While the module docstring mentions logging, it does not provide any user-facing warning, confirmation, or explicit disclosure about network transmission of runtime data and credential use.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The script's user-facing messages are presented in Chinese throughout, including setup prompts, configuration instructions, and completion notes. This imposes a specific language on all users without offering a language choice or documenting that the skill is intended only for a Chinese-speaking or region-specific audience.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · setup.sh (reported line 19)May include surrounding context.

sh
if [ ! -f "$ENV_FILE" ]; then
    cp "$SKILL_DIR/env-example.txt" "$ENV_FILE"
    chmod 600 "$ENV_FILE"
    echo ""
    echo "📝 已创建配置文件:$ENV_FILE"
    echo "请编辑该文件,至少填入:"

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The manifest description is written entirely in Chinese and labels the skill as a meeting-automation flow without indicating that language selection is optional or configurable. Under the policy, forcing a specific language or locale without user opt-in is a natural-language policy concern.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The file includes a Chinese-language comment while the rest of the file is in English, creating an implicit language constraint without offering user choice or explaining a region-specific need. This matches the policy category for language or locale restrictions expressed in natural language.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
97% confidence
Finding

The dependency list uses a lower-bound specifier for requests instead of pinning to an exact version, which makes builds non-reproducible and can silently pull in vulnerable or behavior-changing releases. In a meeting-orchestration skill that likely performs network calls, dependency drift in an HTTP client library increases supply-chain and patch-management risk.

Content

Scanner excerpt · requirements.txt (reported line 1)May include surrounding context.

text
requests>=2.28.0
python-dotenv>=1.0.0
PyYAML>=6.0
pytz>=2023.3

Unverifiable Dependency: requests has 16 known advisory(ies) (CVE-2014-1830 (Exposure of Sensitive Information to an Unauthorized Actor in Requests); CVE-2024-47081 (Requests vulnerable to .netrc credentials leak via malicious URLs); CVE-2024-35195 (Requests `Session` object does not verify requests after making first request wi) +13 more), but the manifest does not pin a version, so it is unknown whether the installed release is affected

Low
Category
Supply Chain
Confidence
95% confidence
Finding

requests has multiple known advisories, and because the manifest does not pin a version there is no way to verify whether deployment will use a patched release. Given this skill's role coordinating meeting creation and likely making outbound API calls, a vulnerable HTTP client could affect credential handling, request verification, or data exposure.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
94% confidence
Finding

python-dotenv is specified with a minimum version only, so installations may resolve to different releases across environments. This weakens reproducibility and can expose the skill to vulnerable or incompatible versions, especially where environment loading affects credentials and service configuration.

Content

Scanner excerpt · requirements.txt (reported line 2)May include surrounding context.

text
requests>=2.28.0
python-dotenv>=1.0.0
PyYAML>=6.0
pytz>=2023.3
python-dateutil>=2.8.2

Unverifiable Dependency: python-dotenv has 2 known advisory(ies) (CVE-2026-28684 (python-dotenv: Symlink following in set_key allows arbitrary file overwrite via ); CVE-2026-28684 (python-dotenv reads key-value pairs from a .env file and can set them as environ)), but the manifest does not pin a version, so it is unknown whether the installed release is affected

Low
Category
Supply Chain
Confidence
87% confidence
Finding

python-dotenv has known advisories and the unpinned requirement prevents verification that a safe version will be installed. Since dotenv libraries can influence secrets and runtime configuration, uncertainty here can contribute to misconfiguration or local file-handling risk.

Content

No source excerpt is available for this finding.