Security audit
Security checks for vulnerabilities and agentic risk
This is a financial-analysis methodology skill collection with disclosed workflows and no evidence of hidden code, exfiltration, persistence, or unsafe automatic actions.
Install only if you want finance-specific analysis templates loaded in OpenClaw. Review outputs before relying on them for lending, investment, insurance, or regulatory decisions, and make sure any configured MCP data connectors expose only the financial documents and providers you intend to use.
The skill description is presented in Chinese and the README consistently instructs users through Chinese-only content, with no indication that another language can be selected. Under the policy, forcing a specific language without user opt-in is a natural-language policy violation unless the locale constraint is explicitly justified.
The manifest description is extremely broad ('use when performing any financial analysis, modeling, due diligence, or report generation'), which can cause the collection to be auto-selected for a very wide range of tasks. In an agent ecosystem, overbroad activation increases the chance that powerful domain instructions are injected into unrelated workflows, leading to unintended behavior, scope creep, or misuse of downstream sub-skills.
Suspicious Unicode normalization or mixed-script content
The primary instructional content begins in Chinese, which implicitly steers the skill toward a specific language/locale. The file does not state that language is optional, user-selected, or required for a region-specific compliance reason, so this appears to violate the language/locale policy.
The file presents core guidance in Chinese while also requiring a specific English announcement and mandatory formatting/output behaviors. This imposes language expectations without explicitly giving the user a language or locale choice, which matches the language/locale policy concern for natural-language policy violations.
The file presents the main skill description and operating instructions in Chinese, while also requiring a fixed English announcement and English-named dependent skills. This effectively imposes language behavior without explicit user opt-in or a documented locale justification.
The file presents core instructional content in Chinese (for example L012 and L018) while also imposing required English phrasing and workflow directives. Because no user opt-in or language preference mechanism is offered, the skill effectively forces a particular language experience, which matches the locale/language policy violation criteria.
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
## Checklist
You MUST create a task for each applicable item and complete them in order:
1. **Parse the request** — Identify deliverable type (pitch book / CIM / comps / DCF / precedent transactions)
2. **Gather source materials** — Collect CIM, financial statements, peer data, market data
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
## Checklist
You MUST create a task for each applicable item and complete them in order:
1. **Parse the request** — Identify deliverable type (pitch book / CIM / comps / DCF / precedent transactions)
2. **Gather source materials** — Collect CIM, financial statements, peer data, market data
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
## Checklist
You MUST create a task for each applicable item and complete them in order:
1. **Parse the request** — Identify deliverable type (pitch book / CIM / comps / DCF / precedent transactions)
2. **Gather source materials** — Collect CIM, financial statements, peer data, market data
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
## Checklist
You MUST create a task for each applicable item and complete them in order:
1. **Parse the request** — Identify deliverable type (pitch book / CIM / comps / DCF / precedent transactions)
2. **Gather source materials** — Collect CIM, financial statements, peer data, market data
The manifest description says to use the skill when finalizing any financial deliverable and before claiming analysis is complete, which is a very broad activation condition. It does not define clear boundaries, exclusions, or concrete trigger phrases, so the skill could be invoked for a wide range of routine finance-related work unintentionally.
The file includes core instructional content in Chinese without indicating that language selection is optional or that the skill is intended only for a Chinese-language context. This can violate language/locale policy because it imposes a language choice on users without opt-in or justification.
Another key instruction line is written in Chinese while the rest of the document is primarily in English, and the file does not explain this mixed-language requirement. That creates an avoidable language constraint without documented user choice or region-specific need.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
|-----------|-----|
| "Revenue grew 15%" without source | "Revenue grew 15% — 10-K, FY2025 vs FY2024, p.38" |
| Using "approximately" without basis | State exact number + source, or mark as estimated |
| Copying numbers from chat/summary without verification | Verify against primary source before using |
| Mixing fiscal year and calendar year without noting | Explicitly state: "FY2025 (ended Jan 2026)" |
| Stale data without date | Always include "as of [date]" |
| Calculation without showing inputs | Show formula and input sources |
The file includes a Chinese-language description while the rest of the manifest is in English, but it does not state whether the skill is intended to operate bilingually or allow the user to choose language. This can create an implicit language preference without clear opt-in or justification.
The skill content switches into Chinese for core instructional text, while the rest of the file is in English, and there is no statement that the user can choose their preferred language. This can violate a language/locale policy when the skill implicitly forces a language without opt-in.
No suspicious patterns detected.