specsfy-aux-stack
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled Node.js script (
scripts/update_stack.mjs) to automate the detection of project technologies. This script is part of the skill's own package. - [SAFE]: File system operations are localized to the project directory. The script reads standard manifest files (
package.json,composer.json) and infrastructure configurations (docker-compose.yml) to identify frameworks and languages. - [SAFE]: The skill does not perform any network requests, nor does it download or execute remote code. All logic is contained within the provided local files.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests data from untrusted project files to populate the
STACK.mdfile, the processing logic is restricted to identifying specific hardcoded package names and does not evaluate or interpret file contents as instructions. The risk of data poisoning is negligible as it uses defined boundary markers (<!-- specsfy:stack:start -->) to manage the documentation block.
Audit Metadata