The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute instructions and helper files from the author's package @nomadamas/k-skill. This allows the skill to load logic dynamically that is not present in the local repository.
[REMOTE_CODE_EXECUTION]: The skill invokes the vendor's CLI via npx -y @nomadamas/k-skill@0, which downloads and runs code from the npm registry at runtime to perform its tasks.
[COMMAND_EXECUTION]: The skill executes a local Python script (scripts/daangn_jobs.py) to perform search and detail retrieval tasks on the Daangn website.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted job posting data from the Daangn website, which could potentially contain malicious instructions intended to influence the agent's behavior.
Ingestion points: The scripts/daangn_jobs.py file fetches JSON and HTML content via fetch_json and fetch_text functions hitting daangn.com endpoints.
Boundary markers: There are no explicit instruction delimiters or 'ignore embedded instructions' warnings defined in the instruction.md to protect the agent when processing the results.
Capability inventory: The skill possesses network request capabilities via Python's urllib.request and shell command execution capabilities via npx.
Sanitization: The script uses html.unescape for basic entity conversion of scraped content but lacks comprehensive sanitization or validation to prevent prompt injection attacks.

daangn-jobs-search