contextloader
Analysis
This is a coherent instruction-only connector for a configured Context Loader API, with clear but sensitive account-scoped knowledge access and a user-directed rebuild action to notice.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
### 创建构建任务 - 方法:`POST` - 路径:`/api/agent-retrieval/in/v1/kn/full_build_ontology`
The documented API surface includes creating a full knowledge-network build job. The examples frame this as user-requested, so it is purpose-aligned, but it is a state-changing/operational action rather than a read-only query.
Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.
从环境变量 `APP_USER_ID` 读取 `x-account-id` ... 调用时固定使用 `x-account-type: app` ... 不传递 `Authorization` 请求头
The skill uses a configured account identifier as an app-scoped request header for the Context Loader service. This is disclosed and purpose-aligned, but it means the skill can access knowledge-network data available to that configured account/service context.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
如果用户没有明确给出 `kn_id`,先读取当前 agent 工作区中的 `SOUL.md` ... 从 `SOUL.md` 的 `## 业务知识网络` 表格中识别候选 BKN 地址,并尝试提取 `kn_id`
The skill intentionally relies on workspace configuration context to select a knowledge network. It includes safeguards against guessing, but stale or incorrect SOUL.md content could still cause the agent to query the wrong knowledge network.
