a1-wayfinder
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill provides structured instructions for 'Charting' and 'Continuing' decision maps. There are no attempts to bypass safety filters, override system instructions, or extract system prompts. The logic focuses on task-specific constraints such as not executing the destination and requiring user confirmation for decisions.
- [DATA_EXFILTRATION]: The skill reads project-specific context files (e.g.,
.agents/marketing-context.md) and writes decision maps to the workspace. It does not perform network operations or hardcode credentials. External links found in documentation are for attribution (GitHub) and support (Telegram), appearing in plain text without malicious intent. - [REMOTE_CODE_EXECUTION]: The skill contains no executable scripts, shell commands, or package dependencies. It relies exclusively on the agent's capability to process and generate markdown content.
- [OBFUSCATION]: Scrutiny of the markdown files and SVG assets revealed no hidden content, Base64 encoding, zero-width characters, or homoglyph substitutions. All instructions and metadata are transparently presented.
Audit Metadata