← Back to Skills Marketplace
Prana七步财务分析
by
goCyberTrade
· GitHub ↗
· v1.0.22
· MIT-0
262
Downloads
0
Stars
0
Active Installs
23
Versions
Install in OpenClaw
/install seven-step-financial-analysis
Description
通过调用 Prana 平台上的远程 agent 完成以下处理：使用7步分析法对中国A股进行全方位财务分析，自动生成包含ROE分解、盈利模式识别、行业对比和投资洞察的HTML报告。支持上海和深圳交易所所有股票。 IMPORTANT: This skill has a mandatory step-by-step p...
Usage Guidance
This skill appears internally consistent with its stated purpose, but review these points before installing or running: 1) Confirm the endpoint: the scripts target https://claw-uat.ebonex.io (a UAT/staging-looking URL). Verify you trust that host and that it is the intended production service. 2) API key handling: the -g flow prints or returns data.api_key and then asks you to set PRANA_SKILL_API_FLAG. Avoid persisting the key on shared or untrusted machines; prefer a temporary session variable if unsure. 3) Logging/exposure: the bundled clients may print JSON including the api_key to stdout — these outputs can end up in terminal history, CI logs, or system log collectors, so capture and store carefully. 4) Data exfiltration surface: using the skill sends task descriptions and presumably financial query inputs to the remote Prana agent (the external server). If your queries contain sensitive info, treat them as transmitted to that service. 5) OpenClaw integration: using the suggested openclaw config command will change agent config — only run with explicit consent. 6) If you need higher assurance, run the bundled scripts locally in a controlled environment, inspect network traffic, and confirm the production/official base URL before setting persistent environment variables. If any of these points are unacceptable, do not persist the key or avoid using the skill.
Capability Analysis
Type: OpenClaw Skill
Name: seven-step-financial-analysis
Version: 1.0.22
The skill bundle is designed to perform A-share financial analysis by interfacing with a remote agent on the Prana platform (claw-uat.ebonex.io). The included scripts (prana_skill_client.js and prana_skill_client.py) are standard API clients that handle key retrieval and task submission. While the SKILL.md instructions guide the AI agent to set persistent environment variables (e.g., via ~/.bashrc or openclaw config), these actions are explicitly tied to user-confirmed steps and are necessary for the skill's stated authentication flow. No malicious intent, data exfiltration, or unauthorized execution logic was detected.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description require calling a remote Prana agent and the skill only requests a single API key (PRANA_SKILL_API_FLAG) and includes client scripts to call the Prana endpoints. Required artifacts (scripts) and network endpoints align with the declared purpose.
Instruction Scope
SKILL.md tightly controls flow: check for PRANA_SKILL_API_FLAG, obtain api_key with bundled client (-g) only after explicit user confirmation, then set PRANA_SKILL_API_FLAG (temp or persistent). The instructions do modify user env (including recommending append to ~/.bashrc or setx) and print/emit API key JSON to stdout when obtained — these are within scope for this skill but are sensitive operations and should be handled with user consent as the doc requires.
Install Mechanism
No install spec; scripts are bundled and executed locally with Node or Python. This is low-risk from an installation standpoint. Note: the default base URL is a UAT-style host (https://claw-uat.ebonex.io) rather than an obviously production Prana domain — worth verifying.
Credentials
Only one required env var (PRANA_SKILL_API_FLAG) is requested and it directly maps to the x-api-key header used by the included clients and endpoints. No unrelated credentials or broad environment access are requested.
Persistence & Privilege
The skill does not force installation or always-on persistence (always:false). It suggests persisting the API key into shell startup or Openclaw config only with user consent. Persisting the API key increases exposure (other processes, backups, shared machines); the guidance to prefer persistence is optional but should be chosen deliberately by the user.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install seven-step-financial-analysis - After installation, invoke the skill by name or use
/seven-step-financial-analysis - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.22
- 修正了持久环境变量设置命令：macOS / Linux（非Openclaw时）从写入 /etc/profile 修改为写入 ~/.bashrc。
- 其它流程规则和行为未发生变动。
- 无新增功能或接口更改。
v1.0.21
No user-visible changes in this version (1.0.21).
- No file changes were detected.
- Behavior and process remain unchanged.
- Documentation and usage instructions are the same as the previous version.
v1.0.20
- API请求地址由 prana.chat 域名切换为 claw-uat.ebonex.io，影响所有相关接口（api-keys、agent-run、agent-result、skill-purchase-history-url）。
- 其它功能及7步分析流程说明无更动，使用体验保持一致。
- 无新功能、无参数/环境变量变动，仅网络请求目标平台发生更新。
v1.0.19
- 增加对 Openclaw 系统的识别与适配，相关检查和环境变量设置命令已单独列出。
- 持久环境变量设置说明中，针对 Openclaw 增加了 `openclaw config set env.PRANA_SKILL_API_FLAG "pkxxxx"`。
- 更新部分持久、临时环境变量命令在 Openclaw、macOS/Linux、Windows 场景下的区分和描述。
- 其余核心流程、确认机制和执行约束规则未发生变动，保持与上一版本一致。
v1.0.18
No user-facing feature changes; environment variable handling instructions updated.
- Updated instructions for setting persistent environment variable on macOS/Linux from editing `/etc/profile` to appending to `~/.zshrc`.
- Enhanced description of required environment variable: `PRANA_SKILL_API_FLAG` must be set before running `-q` or `-r` commands.
- Clarified that obtaining a new api_key is only permitted when explicitly confirmed by the user.
- No impact on API endpoints or analysis process.
v1.0.17
- Updated all API endpoints to use the new prana.chat domain instead of claw-uat.ebonex.io.
- No user-facing functional changes; all descriptions and usage flows remain unchanged.
- Ensures continued compatibility and data access via the new API server addresses.
v1.0.16
- Migrated all API endpoints from "claw-uat.ebonex.io" to the production "prana.chat" domain.
- Streamlined and clarified the environment variable setup: scripts now require `PRANA_SKILL_API_FLAG` to be set before execution and do not auto-fetch keys.
- Updated process and documentation to reflect that all API key acquisition should be confirmed with the user and set via `openclaw` if persistent storage is desired.
- Simplified skill usage instructions and step-by-step guidance, focusing on fast, compliant A股 financial analysis.
- Changelog summary: Production rollout, stricter environment requirements, clearer user consent process, and updated usage steps.
v1.0.15
Version 1.0.15
- Clarified language in the skill description and process steps to improve readability (especially around step-by-step flow and user permissions).
- Changed emphasis from "DO NOT" (all caps) to "Do not" (sentence case) and revised key warnings for international/English audiences.
- Added/enhanced notes that each api_key is unique and must not be re-fetched or rotated without user authorization.
- No changes to core logic or API endpoints—process enforcement only clarified, not altered.
v1.0.14
- Enhanced and clarified the step-by-step process for acquiring and configuring the `PRANA_SKILL_API_FLAG` environment variable, emphasizing mandatory user confirmation and the prohibition of skipping or merging steps.
- Added strict rules preventing key re-fetching, environment variable overwrites, or process shortcutting under all circumstances.
- Revised documentation with system-specific command examples, detailed user confirmations, and clear action vs. confirmation breakpoints.
- Updated usage scenarios, script parameter notes, and output presentation guidelines for greater consistency and user safety.
- Specified the distinction between temporary and persistent environment variable settings and their respective risks/advantages.
- All logic and workflow are now fully aligned to ensure auditability, security, and compliance with Prana platform requirements.
v1.0.13
Version 1.0.13
- 更新说明场景描述，使其更突出“快速诊断A股公司财务健康度与盈利模式”。
- 优化运行流程说明，明确脚本调用时不允许用命令行方式设置环境变量。
- 明确脚本参数统一使用 `-q` 用于传递分析需求。
- 补充了关于脚本输出（包括链接展示、预期 JSON 返回、尝试上限处理）的用户提示规范。
- 无代码文件变更，仅文档细节修订与流程规范完善。
v1.0.12
- Skill renamed为“A股财务分析”，并明确聚焦中国A股全方位财务分析（支持沪深交易所全部股票）。
- skill 描述优化，明确报告内容包括ROE分解、盈利模式、行业对比和投资洞察，并生成HTML报告。
- 更严格规范环境变量获取和使用，新增“required_env_note”，并强调运行前必须已设置 PRANA_SKILL_API_FLAG，脚本本身不再自动请求 token。
- 增加安全合规要求：禁止在命令行直接传递密钥环境变量，须先设置后执行。
- 明确获取历史记录接口只能由用户主动请求时才允许调用，避免自动拉取。
- 输出网页链接结果时，要求展示为用户可直接点击的形式，提高用户体验。
v1.0.11
Version 1.0.11
- 文档中的“若用户不同意写入 OpenClaw 全局配置，应停止本技能流程处理”修改为“应停止本程序流程处理”
- 对 thread_id 说明进行了优化，首次调用需传空；续用同一对话时传递上一次 agent-run 返回结果中的 thread_id
- 删除了旧版关于 thread_id “技能维度维护”与示例说明，改以“首次为空，后续用返回值”描述
- 其他描述表述略作调整，实际功能未变
v1.0.10
- Added a clarification regarding OpenClaw global environment variable persistence as an optional (可选) step, with an explicit “补充说明” note.
- No changes to core logic or interfaces. All functionality remains unchanged.
- General documentation refinement for improved clarity on environment variable handling.
v1.0.9
- Clarified that the environment variable PRANA_SKILL_API_FLAG should not be globally persisted without explicit user consent.
- Added instruction to stop the process if the user does not agree to global configuration.
- No logic or technical flow changes; update is focused on strengthening user privacy and consent requirements.
v1.0.8
- 简化了api_key环境变量配置流程，去除“需用户明确同意”与“可选全局持久化”相关说明
- 明确要求通过OpenClaw专属命令设置PRANA_SKILL_API_FLAG环境变量，确保全局生效
- 其余七步财务分析流程、接口调用及用法保持不变
v1.0.7
- 技能名称与描述更新，明确为“Prana七步财务分析”，突出七步分析法及指定股票应用场景。
- 调整认证流程，新增明确“用户同意”与“全局持久化环境变量”可选步骤。
- 修改API获取方式：接口 `api/v2/api-keys`请求方式由 POST 改为 GET。
- 优化脚本执行与 thread_id 管理，细化多技能会话的 thread_id 使用规则。
- 示例和指令更新，更贴合七步财务分析常见用法与实际需求。
v1.0.6
- Skill名称由“Prana七步财务分析”改为“A股财务分析”，描述更简洁，强调A股财务分析功能。
- 使用场景示例由具体股票分析变为泛化的A股财务分析任务和建议，适用范围更广。
- 说明部分简化，去除七步流程等详细步骤，脚本参数介绍更通用。
- 线程管理(thread_id)的描述更为简明，区分首次与续会话调用。
- 其他结构和接口调用方式保持一致，细节表述更简化明了。
v1.0.5
- 调整 thread_id 维度为“技能级别”，同一 OpenClaw 会话中不同技能分别独立维护 thread_id，上次每次调用都需按技能分别传递。
- 详细补充了 thread_id 传递的规则和示例，增强对多技能并发场景的支持。
- 其余功能、接口和使用流程保持不变。
v1.0.4
Version 1.0.4 of seven-step-financial-analysis has no file changes from the previous release.
- No updates or modifications detected in this version.
- Functionality, documentation, and configuration remain identical to the previous release.
v1.0.3
- No user-facing changes; content and configuration remain the same as the previous version.
- Version bump without modifications to functionality, documentation, or network requests.
Metadata
Frequently Asked Questions
What is Prana七步财务分析?
通过调用 Prana 平台上的远程 agent 完成以下处理：使用7步分析法对中国A股进行全方位财务分析，自动生成包含ROE分解、盈利模式识别、行业对比和投资洞察的HTML报告。支持上海和深圳交易所所有股票。 IMPORTANT: This skill has a mandatory step-by-step p... It is an AI Agent Skill for Claude Code / OpenClaw, with 262 downloads so far.
How do I install Prana七步财务分析?
Run "/install seven-step-financial-analysis" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Prana七步财务分析 free?
Yes, Prana七步财务分析 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Prana七步财务分析 support?
Prana七步财务分析 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Prana七步财务分析?
It is built and maintained by goCyberTrade (@gocybertrade); the current version is v1.0.22.
More Skills