The Agent Skills Directory

[SAFE]: The skill implements a multi-agent documentation pipeline (scanner, extractor, profiler, writer, checker, critic) designed to analyze local codebases. The orchestration is transparent and follows documented logic.
[SAFE]: Internal Bun scripts (manifest-sync.ts, scaffold-eval-loop.ts, append-loop-result.ts) contain explicit security controls to prevent path traversal and symbolic link attacks. For example, append-loop-result.ts uses realpathSync and checks that artifacts reside within the project root.
[COMMAND_EXECUTION]: The skill utilizes local command execution for internal synchronization and scaffolding tasks. These executions are constrained to the skill's own internal script files and do not allow for the execution of arbitrary or user-supplied shell commands.
[DATA_EXFILTRATION]: No evidence of unauthorized data transmission or network activity was discovered. The skill operates purely on local project files to derive documentation content.

write-docs