simple-changelogs-web-cms
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a well-documented and professionally structured utility for changelog management. All included scripts operate locally within the repository to manage state and provide tooling for maintainers and agents. No indicators of obfuscation, exfiltration, or persistence were found.
- [COMMAND_EXECUTION]: The skill's bundled scripts (
scripts/setup.ts,scripts/check-fork-sync.sh) execute localgitcommands via subprocesses to inspect repository history, verify commit provenance, and identify project structure. These operations are standard for development tools and do not involve executing commands from untrusted remote sources. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-controlled Markdown files (
CHANGELOG.mdandDEVELOPER_CHANGELOG.md) during history queries and release reconciliation. This ingestion surface is mitigated by a deterministic, structural parser (scripts/lib/changelog-parse.ts) that extracts entries based on specific Markdown delimiters, preventing the interpretation of arbitrary natural language as instructions. - Ingestion points: Processes
CHANGELOG.mdandDEVELOPER_CHANGELOG.mdvia thescripts/query.tsCLI. - Boundary markers: Uses structural Markdown elements like versioned headings and bullet points to delimit entries.
- Capability inventory: The skill is capable of writing to repository-local configuration files (
.simple-changelogs.json) and executing local Git commands for repository state inspection. - Sanitization: Implements character escaping for agent identity strings when generating signature comments to ensure Markdown integrity.
Audit Metadata