release-changelog-discord-message

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external content by reading release changelogs and issue comments to synthesize a release announcement.
  • Ingestion points: Reads the release changelog file (releases/vYYYY.MDD.P.md) and the release issue thread (comments, linked issues, and recent commits).
  • Boundary markers: Absent. No specific delimiters or boundary guards are instructed to isolate embedded text from structural instructions.
  • Capability inventory: Performs structural internal API updates (PUT /api/issues/{releaseIssueId}/documents/discord_announcement and POST /api/companies/:companyId/cases) to publish the announcement text within the issue management platform. No subprocess creation, direct shell execution, or arbitrary outbound network tools are invoked.
  • Sanitization: Absent. The instructions do not prescribe sanitization or escaping filters for input text before formatting the output announcement.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 03:47 PM
Security Audit — agent-trust-hub — release-changelog-discord-message