Back to skill

Security audit

HK IPO Profile and Watchlist Manager

Security checks for vulnerabilities and agentic risk

Overview

The skill is presented as a profile and watchlist manager, but its bundled runtime exposes broader IPO data, scoring, review-history, suggestion-application, network, and persistence capabilities.

Install only if you are comfortable with a bundled HK IPO decision-support CLI, not just a local profile/watchlist helper. Review the broader commands before use, run it in a minimally privileged environment, avoid exposing unrelated secrets, and prefer a reviewed lockfile for dependencies.

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
runtime/hkipo-next/pyproject.toml:11
Finding

Unbounded Runtime Dependencies Without a Bundled Lockfile

Content
View full analysis
=4.14.3", "httpx>=0.28.1", "lxml>=6.0.4", "opencc-python-reimplemented>=0.1.7", "pydantic>=2.13.0", "requests>=2.33.1", ] ``` The documented execution workflow in `SKILL.md:20` uses: ```bash uv run --directory runtime/hkipo-next hkipo-next ... ``` ### Technical Analysis All runtime dependencies specify only minimum versions. No reviewed lockfile is included in the audited project. Consequently, `uv run` can resolve versions that were published after this Skill was reviewed. This does not establish that any current dependency is malicious. However, it creates a supply-chain vulnerability because the code ultimately installed and imported at runtime is not restricted to the versions represented by the audited package. A compromised dependency release, compromised package-publisher account, malicious transitive dependency, or unexpectedly incompatible future release could become part of the effective runtime without any modification to the Skill itself. Python packages execute code with the invoking user's privileges when their modules are imported. Depending on build and installation behavior, package-controlled code may also execute during package construction or installation. ### Attack Path 1. An attacker compromises the publishing account or release infrastructure of a declared dependency or one of its transitive dependencies. 2. The attacker publishes a new version satisfying the applicable open-ended `>=` constraint. 3. A user follows the documented `uv run` workflow in an environment without a previously verified and immutable dependency resolution. 4. The resolver selects and installs the newly published version. 5. The compromised package executes when installed or importe ...[truncated 1084 chars]
Remediation
View remediation
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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
Findings (166)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Across the findings, the strongest consistent signal is that the documented purpose is narrow profile/watchlist management, while the referenced runtime reportedly exposes substantially broader behavior such as market-data retrieval, scoring, review history, parameter management, caching, and external network access. This is dangerous because users and policy engines may grant trust based on the declared purpose, while the actual runtime can perform unrelated and more sensitive actions.

Content

No source excerpt is available for this finding.