← 返回 Skills 市场
prana-astock-financial-analysis
作者
goCyberTrade
· GitHub ↗
· v1.0.19
· MIT-0
251
总下载
0
收藏
1
当前安装
20
版本数
在 OpenClaw 中安装
/install prana-astock-financial-analysis
功能描述
通过调用 Prana 平台上的远程 agent 完成以下处理：分析A股上市公司财务状况,从6个维度展示(盈利能力、偿债能力、营运能力、成长能力、现金流质量、估值水平),生成交互式HTML报告,默认分析近8个季度数据。 IMPORTANT: This skill has a mandatory step-by-st...
安全使用建议
This skill appears to do what it claims: it uses a single API key (PRANA_SKILL_API_FLAG) to call a remote Prana agent and returns an analysis report. Before installing/using: (1) Confirm you trust the remote host (https://claw-uat.ebonex.io appears to be a UAT/staging domain); (2) Do not paste the full api_key into chat—follow the SKILL.md process and give explicit consent before retrieving/setting the key; (3) Prefer setting the key as a temporary/session env var unless you intentionally want persistent reuse; (4) Be aware that choosing the global env option will persist the key for future runs (consider key rotation and who has access to your environment); (5) If PRANA_SKILL_API_FLAG already exists, follow the skill rule and do not re-fetch/overwrite without an explicit user decision. If you need greater assurance, ask the skill author for the official production endpoint, key lifetime/rotation policy, and a vendor identity you can verify.
功能分析
Type: OpenClaw Skill
Name: prana-astock-financial-analysis
Version: 1.0.19
The skill bundle is a legitimate client for a financial analysis service hosted on claw-uat.ebonex.io. The provided Python and Node.js scripts (prana_skill_client.py/js) are standard API wrappers that send user queries and poll for results using an API key stored in the PRANA_SKILL_API_FLAG environment variable. While SKILL.md contains highly prescriptive instructions for the AI agent regarding API key management and the configuration of global environment variables, these instructions include mandatory user confirmation steps and are consistent with the operational requirements of the service. No evidence of data exfiltration, unauthorized execution, or malicious intent was detected.
能力标签
能力评估
Purpose & Capability
Name/description (A-share financial analysis) align with the included client scripts and network calls. The only required secret is PRANA_SKILL_API_FLAG which is needed to call the remote agent endpoints; no unrelated credentials or system accesses are requested.
Instruction Scope
SKILL.md imposes a strict, multi-step process around acquiring and setting an api_key and instructs running the included client scripts. The steps focus on user confirmation and env var setting and do not ask the agent to read unrelated files or secrets. The strict prohibition on re-fetching keys if PRANA_SKILL_API_FLAG exists is unusual but consistent with the described workflow.
Install Mechanism
No install spec; this is an instruction-only skill with two simple client scripts included. Nothing is downloaded from external, arbitrary URLs and no installers are executed—low install risk.
Credentials
Only PRANA_SKILL_API_FLAG is required. That single API key is proportionate to calling the remote Prana endpoints; no other secrets, tokens, or config paths are requested.
Persistence & Privilege
always:false and the clients are not autonomously modifying other skills. SKILL.md recommends (user-choice) writing PRANA_SKILL_API_FLAG as a global env via an 'openclaw config set' command; that would create persistent credential state if the user chooses it—this is a valid option but increases persistence and should be chosen consciously by the user.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install prana-astock-financial-analysis - 安装完成后，直接呼叫该 Skill 的名称或使用
/prana-astock-financial-analysis触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.19
**Changelog for prana-astock-financial-analysis v1.0.19**
- Strengthened environment variable handling rules: Once PRANA_SKILL_API_FLAG is set, all forms of api_key refetch/refresh (GET /api/v2/api-keys) are strictly and absolutely forbidden.
- Updated process rules: No key rotation, overwrite, or re-fetch is permitted under any scenario if PRANA_SKILL_API_FLAG already exists; failures must be reported to the user.
- Clarified output and retry rules: Added explicit requirement for user confirmation before continuing after error messages or “retries/attempts reached” prompts.
- Improved scenario description to cover both financial health and valuation analysis.
- Documentation streamlined for clarity; repeatedly emphasizes strict one-way/one-time key issuance behavior and prohibition of unsanctioned retries or self-initiated key fetches.
v1.0.18
No user-facing changes in this version.
- No file changes detected.
- Skill behavior and processes remain unchanged.
v1.0.17
No changes detected in this version.
- There are no file changes compared to the previous release.
- Skill behavior, step-by-step process, and usage instructions remain unchanged.
v1.0.16
- Clarified that each api_key from GET /api/v2/api-keys is independent and unique; you must not re-fetch or replace keys without explicit user permission.
- Added additional warnings: never re-pull or re-issue a key on your own initiative, even to “try another one”—always use the configured value unless the user explicitly authorizes otherwise.
- Updated process rules and prohibitions to reinforce the uniqueness and one-time issuance of api_key, and the need for strict user confirmation on any reconfiguration.
- No behavioral or functional changes to API or core flow; documentation and enforcement guidance only.
v1.0.15
**Summary**: Adds a strict rule prohibiting overwriting or rotating the PRANA_SKILL_API_FLAG environment variable without explicit user consent if it is already set.
- Introduced a mandatory check: if PRANA_SKILL_API_FLAG is already present, do not re-fetch or overwrite the key without explicit user approval.
- Updated process rules to clarify that Step 2 (fetching and setting the key) cannot be performed again unless the user clearly requests it.
- Enhanced error handling: do not auto-refresh environment variables or keys on errors; always defer to user instructions.
- Clarified documentation and instructions to ensure compliance with the new strict environment variable handling policy.
v1.0.14
- Step 2 流程支持用户自主选择将 api_key 设置为「当前会话临时环境变量」或「全局环境变量」（之前仅允许全局设置）。
- 在用户确认时，需说明两种设置方式的优缺点，由用户决定。
- 流程与规则描述已对应更新为支持会话级环境变量路径（如 PowerShell `$env:VAR=...` 或 cmd `set VAR=...`）。
- 所有自动取钥与环境变量配置动作，依然必须先获用户明确同意（流程刚性要求未改）。
- 其余流程保持不变，无新增接口或参数变动。
v1.0.13
- Strengthened Step 2 process rules to require user confirmation and global environment variable write before api_key retrieval; Step 2 can never be executed alone or out of order.
- Clarified that checking/setting PRANA_SKILL_API_FLAG must follow strict sequence: explicit confirmation → GET api_key → set global env var.
- Prohibited any session-only or partial api_key operations; global write via openclaw config is now the only allowed method.
- Expanded and emphasized warnings and forbidden patterns for Step 2, ensuring no bypass or partial execution regardless of context.
- Updated step numbering: all steps after Step 2 incremented (e.g., former Step 4 is now Step 3), with rules adapted accordingly.
- No changes to API behavior, core functionality, or intended usage—this release is focused on stricter process compliance and user consent.
v1.0.12
**Major update: Enforces a strict, step-by-step process for credential handling and environment configuration.**
- Added explicit step-by-step rules requiring user confirmation before any credential or config actions.
- All critical steps (e.g., obtaining and writing api_key) must announce intent and await user approval where marked.
- Clarified that no steps may be skipped or merged without explicit user consent.
- Enhanced output presentation guidelines for URLs and service messages.
- Disallowed any automatic or "helpful" retrieval of history/record links unless explicitly requested by the user.
v1.0.11
**重要更新：简化并优化A股财务分析助手，增强脚本运行和权限安全要求。**
- skill 说明和使用场景简化，明确6大财务分析维度和默认分析近8季度。
- 要求在脚本运行前必须手动设置环境变量 PRANA_SKILL_API_FLAG，客户端脚本不再自动获取或配置 api_key。
- 环境变量写入需事先获得用户明确同意，不得自动持久化用户密钥。
- 精简流程，不再强制维护 thread_id 文件或会话上下文，无需特定命名存储。
- 严格禁止通过命令行设置临时环境变量；必须先全局配置再运行脚本。
- 说明“历史记录”接口仅在用户明确需求下调用，返回链接需用户可直接点击。
- 增加结果输出、JSON处理和重试提示的具体用户交互规范。
v1.0.10
Version 1.0.10
- Skill名称由“A股财务分析”改为“A股财务分析助手”，描述简化。
- 新增会话 thread_id 持久化及 workspace 目录读写规范，明确需用 session_id 目录保存/读取 thread_id。
- 更新环境变量配置说明，要求全局持久化并用 OpenClaw 专属命令设置。
- 规范脚本参数为 question 和 thread_id，详细说明调用顺序与流程。
- 历史请求（技能获取记录）查询流程简化，强调直接输出结果给用户，避免日记和存储泄露。
- 精简使用场景与接口说明，去除步骤细节和部分约束表述。
v1.0.9
prana-astock-financial-analysis 1.0.9
- 优化了脚本参数说明，将“生成宁德时代盈利能力与现金流质量的交互式报告”统一为“A股财务分析交互式报告”。
- 增加了输出脚本结果时对预期内 JSON 返回及达到尝试次数上限的用户指引，明确由用户决定下一步操作。
- 补充并细化了链接输出格式要求，强化必须使用可点击链接的用户体验指引。
- 略微简化了 Node.js 和 Python 3 脚本参数说明，并去除了 thread_id 续写相关说明。
- 仅修改文档，无功能及接口变更。
v1.0.8
No user-facing changes detected in this version.
- No file changes were made compared to the previous version.
- No updates to SKILL.md or feature set.
v1.0.7
prana-astock-financial-analysis 1.0.7
- No file changes detected in this release.
- Functionality and documentation remain unchanged from the previous version.
v1.0.6
Version 1.0.6
- 更新技能功能描述，明确六大财务分析维度并强调交互式HTML报告输出，默认分析近8个季度数据。
- 精简和优化使用场景说明，更聚焦A股财务健康诊断，增加具体示例。
- 修改运行环境变量检查建议，推荐使用 `env | grep PRANA_SKILL_API_FLAG`。
- 强化脚本安全要求，禁止在命令行中直接写密钥，确保环境变量安全设置。
- 保持核心API流程和安全合规指引未变。
v1.0.5
Version 1.0.5
- No file changes detected in this release.
- Behavior and documentation remain identical to previous version.
- No user action required.
v1.0.4
Summary: Environment variable usage clarified; skill now requires PRANA_SKILL_API_FLAG to be pre-set before running client scripts.
- 新增 required_env_note，明确需预先配置 PRANA_SKILL_API_FLAG，并说明随附脚本不会自动请求 api-key。
- 审计说明第四项新增对 required_env 的运行要求及取钥流程的详细补充。
- 其他核心流程未做更改，接口行为和调用方式保持一致。
- 文档进一步强调了环境变量缺失时需由调用方先行设置。
v1.0.3
- Skill名称优化为“A股财务分析”，描述和使用场景更加通用。
- 明确要求在写入全局环境变量前必须获得用户的明确同意。
- 新增历史请求地址接口调用的用户授权和触发原则，避免非指令性自动拉取。
- 强调向用户展示URL时应提供易于直接点击的链接格式，提升用户体验。
- 更新接口方法，GET请求用于获取API key；清晰区分各接口用途和调用时机。
v1.0.2
- Improved thread_id management: now maintained per skill, enabling conversation context retention for each skill independently.
- Updated documentation in SKILL.md to clarify thread_id usage and calling patterns.
- Removed obsolete instructions about reading thread_id from files; now based on last agent-run result per skill.
- No changes to API endpoints or environment variables.
v1.0.1
Version 1.0.1
- No file changes detected in this release.
- No updates to features, code, or documentation.
v1.0.0
A股财务分析助手 1.0.0
- 首次发布，支持通过Prana平台代理远程分析指定A股股票的财务能力（如盈利能力、营运能力）。
- 自动管理api_key环境变量，按需从Prana平台获取并全局持久化。
- 支持Node.js优先、本地Python备选两种客户端执行模式。
- 提供获取技能历史请求记录的专用接口及安全说明。
- 明确约束外部接口调用，仅限于claw-uat.ebonex.io域名下。
元数据
常见问题
prana-astock-financial-analysis 是什么？
通过调用 Prana 平台上的远程 agent 完成以下处理：分析A股上市公司财务状况,从6个维度展示(盈利能力、偿债能力、营运能力、成长能力、现金流质量、估值水平),生成交互式HTML报告,默认分析近8个季度数据。 IMPORTANT: This skill has a mandatory step-by-st... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 251 次。
如何安装 prana-astock-financial-analysis？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install prana-astock-financial-analysis」即可一键安装，无需额外配置。
prana-astock-financial-analysis 是免费的吗？
是的，prana-astock-financial-analysis 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
prana-astock-financial-analysis 支持哪些平台？
prana-astock-financial-analysis 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 prana-astock-financial-analysis？
由 goCyberTrade（@gocybertrade）开发并维护，当前版本 v1.0.19。
推荐 Skills