write-adr
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends using
npx ctx7@latestor thectx7CLI to fetch technology documentation. This instruction causes the agent to download and execute an external package from the NPM registry during the documentation gathering phase. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process local project documentation (
docs/PRD.md) which may contain untrusted content that could influence the agent's behavior. - Ingestion points: The agent is instructed to read
docs/PRD.mdor user-provided feature briefs to extract system requirements. - Boundary markers: The instructions do not include specific delimiters or warnings to the agent to ignore control sequences or malicious instructions that might be embedded within the PRD files.
- Capability inventory: The skill grants the agent the ability to write multiple files to the local file system (
docs/ADR/) and execute shell commands via thectx7documentation tool. - Sanitization: There is no evidence of sanitization or validation performed on the ingested PRD content before it is used to generate the architectural documents.
Audit Metadata