The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or the associated script. The code uses standard libraries and performs purely mathematical operations.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data through the processing of publication lists.
Ingestion points: Publication list data in JSON format supplied by the user as described in SKILL.md and processed by scripts/score_docentlik.py.
Boundary markers: Absent in the raw prompt input, but the skill enforces a specific data schema through strict JSON parsing in the scoring script.
Capability inventory: The skill utilizes file system permissions (Read, Write, Edit) and shell access (Bash) to execute its local Python calculation script.
Sanitization: Input is sanitized by a deterministic Python script that validates JSON types and performs arithmetic calculations. The script does not use eval() or string interpolation into shell commands, and the skill's instructions prohibit the model from asserting final eligibility based on the processed data.

alterlab-docentlik-eligibility