team-brief
Warn
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTIONCREDENTIALS_UNSAFEDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands and dynamic Python snippets to generate JSON payloads and perform API interactions with HackMD and other services.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. It ingests untrusted content from GitHub, Slack, and Linear and incorporates it into summaries and alerts without explicit boundary markers or sanitization logic, potentially allowing external content to influence the agent's output.
- [CREDENTIALS_UNSAFE]: The skill is designed to read an API token from a local file path at
.claude/settings/hackmd-tokenfor authentication purposes. - [DATA_EXFILTRATION]: Aggregated team data, including discussion threads and meeting notes, is transmitted to the external service
api.hackmd.io. HackMD is recognized as a well-known documentation service.
Audit Metadata