strategic-compact
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation provides instructions for manual installation that involve adding a shell command to the agent's configuration (
~/.claude/settings.json). This command executes a local Node.js script (suggest-compact.js) as a hook duringEditandWriteoperations. - [DATA_EXFILTRATION]: The skill logic, as described, monitors session metadata such as
usagerecords (token counts) andtool_callsfrom the local session transcript. This information is used locally to trigger user-facing suggestions for context management and does not involve sending data to external servers. - [EXTERNAL_DOWNLOADS]: The documentation references external resources including a social media link for context optimization guides and mentions a
token-optimizerMCP tool. No automated or hidden downloads are performed by the skill itself.
Audit Metadata