Anima Backend exposes 11 tools with basic schemas and descriptions, but falls short of production quality. All tools have descriptions (10-200 chars, within baseline range) and input schemas with type definitions. However, parameter descriptions are sparse or missing entirely. Critical gaps: run_shell lacks any input validation guidance; fetch_url has no guidance on URL format or response size limits; file operations (read_file, write_file) lack path traversal protection hints. No error handling guidance across any tool. Tool names are clear (verb_noun pattern), but descriptions are minimal and lack WHEN/WHY context. No output schemas documented. Composition is reasonable (single responsibility per tool), but several tools (install_skill, remove_skill) perform destructive operations without confirmation patterns or dry-run support.

• Document output schemas for all tools. Example: list_skills should return {skills: [{name: string, description: string}], total: number}. This enables LLM to chain calls and extract data.
• Add parameter descriptions to all tools. fetch_url.strip_html has a description, but fetch_url.url does not. Every parameter needs a description explaining format, constraints, and examples.
• Add error recovery guidance to run_shell: 'If command fails, check syntax and permissions. Common errors: command not found, permission denied, timeout (>30s).'
• Implement confirmation pattern for destructive operations. remove_skill should require a confirm parameter or return a confirmation token that must be passed in a follow-up call.
• Add path traversal protection hints to read_file and write_file: 'Paths are sandboxed to the files directory. Relative paths like ../../../etc/passwd are rejected.'

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry