fetch_url
A built-in URL fetching tool in CodeSmith. Its description steers the model away from using the curl command, encouraging use of the dedicated tool for retrieving web content instead.
Timeline (last 90 days)
CodeSmith external content source tag policy: fetch_url uses the URL hostname, web_search uses a fixed tag, MCP tools take a server-name fragment, and github_* is uniformly recorded as github.com; tags are byte-stable, not timestamps or request IDs.
CodeSmith's tool descriptions contain about ten anti-pattern lists in the form 'Use this instead of ...', for example write_file discouraging heredoc, edit_file discouraging sed -i, fetch_url discouraging curl, and grep_files discouraging grep -r.
Largha-150M-v1 is weak at tool routing: the model card's case shows it returned fetch_url instead of web_search at about 0.33 confidence.
All Facts (3)
CodeSmith external content source tag policy: fetch_url uses the URL hostname, web_search uses a fixed tag, MCP tools take a server-name fragment, and github_* is uniformly recorded as github.com; tags are byte-stable, not timestamps or request IDs.
50%UnverifiedCodeSmith's tool descriptions contain about ten anti-pattern lists in the form 'Use this instead of ...', for example write_file discouraging heredoc, edit_file discouraging sed -i, fetch_url discouraging curl, and grep_files discouraging grep -r.
50%UnverifiedLargha-150M-v1 is weak at tool routing: the model card's case shows it returned fetch_url instead of web_search at about 0.33 confidence.
50%