byte-codebase-harness
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary function is to generate repository documentation and configuration files (e.g., CLAUDE.md, AGENTS.md, .claude/settings.json) to aid AI agents. All operations are local and consistent with the stated purpose of codebase mapping.
- [COMMAND_EXECUTION]: The workflow involves repository inspection using local utilities such as
rg(ripgrep) and file system reads to extract information from build files, package managers, and directory structures. - [PROMPT_INJECTION]: The skill ingests untrusted repository content, which creates a surface for indirect prompt injection.
- Ingestion points: The skill reads existing
CLAUDE.md,AGENTS.md, and project configuration files from the repository root. - Boundary markers: There are no explicit instructions or delimiters used to prevent the agent from following malicious instructions that might be embedded in the files it reads.
- Capability inventory: The skill possesses extensive file-writing capabilities to the repository root and the
.byte-os/directory, and it executes shell commands for repository searching. - Sanitization: No explicit sanitization or validation of the content read from external repository files is performed before it is used to generate new artifacts.
Audit Metadata