The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python utility, scripts/check_invariants.py, to perform deterministic validation of academic content. This script uses standard Python libraries (re, json, argparse) and does not perform network operations, access sensitive system files, or install external dependencies.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided academic prose, which constitutes an ingestion point for untrusted data.
Ingestion points: Text passages, abstracts, and author writing samples provided by the user in the agent's context.
Boundary markers: The skill uses a 'preservation ledger' and specific 'output contracts' to delimit and verify facts independently of stylistic changes.
Capability inventory: The skill can execute the local check_invariants.py script via a shell command and read/write the case.json file used for verification.
Sanitization: The Python script treats input as literal strings for regex-based token matching, and the instructions prioritize scholarly fidelity over style, reducing the likelihood of the agent following instructions embedded in the user's text.

academic-humanizer