The Agent Skills Directory

[SAFE]: The skill provides comprehensive and legitimate documentation for the jq utility. All provided bash commands are standard examples for JSON manipulation (filtering, mapping, object/array operations).
[COMMAND_EXECUTION]: The skill uses allowed-tools: Bash(jq *) and other standard tools (cat, Grep, Glob). The commands listed in the documentation (e.g., jq '.' file.json, jq -r '.email') are strictly limited to the intended purpose of the skill and do not include dangerous flags, shell escapes, or arbitrary command execution patterns.
[EXTERNAL_DOWNLOADS]: The skill includes a curl example (curl -s https://api.example.com/data | jq '.results'). This is documented as a generic placeholder for API processing and targets a non-existent example domain. Per [TRUST-SCOPE-RULE], such generic placeholders in a documentation context are safe and do not escalate the verdict.
[REMOTE_CODE_EXECUTION]: No remote code execution patterns or piped-to-shell downloads were detected.
[DATA_EXFILTRATION]: There are no patterns of sensitive file access (like .ssh or .env) combined with network operations. The network examples are one-way reads (GET) from placeholder APIs.
[OBFUSCATION]: No obfuscated strings, Base64 encoding, zero-width characters, or homoglyphs were found in the markdown or JSON files.

jq-json-processing