The Agent Skills Directory

[SAFE]: The skill implements text analysis functionality using standard Python libraries. No malicious patterns such as obfuscation, credential harvesting, or remote code execution were detected.
[DATA_EXFILTRATION]: The script scripts/analyze_text.py possesses the capability to read content from the local filesystem via the --file parameter. This is an intended feature for document analysis and is not accompanied by any network capabilities that would enable exfiltration.
[PROMPT_INJECTION]: The skill handles untrusted data which could potentially contain indirect prompt injections.
Ingestion points: Text is ingested from user prompts or local files in scripts/analyze_text.py.
Boundary markers: No specific boundary markers are used to isolate content during processing.
Capability inventory: The skill is limited to file reading and standard output. It cannot write to the filesystem, access the network, or execute shell commands.
Sanitization: While no sanitization is performed on the input, the output is transformed into statistical counts and word frequencies, which effectively mitigates the risk of the agent executing instructions embedded in the analyzed text.

text-analyzer