critique-argument
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for the legitimate purpose of reviewing argumentative structure (claims, grounds, warrants, etc.) in documents like essays and proposals. It provides clear criteria and a multi-pass protocol to ensure consistency.
- [EXTERNAL_DOWNLOADS]: The skill mentions a prerequisite to install
jsonschema. This is a well-known, standard library used for validating JSON against a schema, which is consistent with the skill's goal of emitting contract-valid JSON findings. - [COMMAND_EXECUTION]: The skill executes two local scripts:
scripts/checks.pyandscripts/merge.py. These scripts perform deterministic measurements (word counts, sentence ratios, lexicon matching) and arithmetic operations for envelope assembly. The code analysis shows no unauthorized network access, sensitive file reads, or dynamic code execution. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied prose. It includes specific protocol steps for 'Clean-context critique' to mitigate bias or manipulation from the input text by disregarding authorial framing and using a defined 'target reader' construct.
Audit Metadata