climike

7 ms·

1.
by climike 5mo ago
Exactly! Number of turns, average tokens to achieve a task using your CLI, as well as average number of characters being returned per CLI command alongside other metrics: all important to both users and agents! I am working on allowing to a
2.
by climike 5mo ago
Working on testing and monitoring agent-readiness of CLIs with www.cliwatch.com, some interesting challenges in building test suites and analyzing data :)
3.
by climike 6mo ago
Also worth pushing for a more standardized skills command for CLIs, similar to —help, but for (agent/human) workflows, https://cliwatch.com/blog/designing-a-cli-skills-protocol (if you ship these with your CLI, yo
4.
by climike 6mo ago
Resource allocation based on your hackernews upvotes? Thanks in advance folks ;)
5.
by climike 6mo ago
See also https://cliwatch.com/blog/designing-a-cli-skills-protocol
6.
by climike 6mo ago
We are working on supporting agent harnesses @ www.cliwatch.com, so both 1. LLM model as well 2. LLM model + harness performance can be evaluated against your software/CLI. We also support building evals against your doc suite. End res
7.
by climike 7mo ago
Building www.cliwatch.com, so you can keep an eye on how agent-friendly your CLI is ;) feel free to request a benchmark against your CLI docs. Cheers
8.
by climike 7mo ago
cliwatch.com, creating some benchmarks, reach out if you are interested :)
9.
by climike 7mo ago
In a similar fashion it appears that article was automated - did the author read every word in their own article?
10.
by climike 7mo ago
Not sure about the end of thinking, would say that this is the start of managing ever more stochastic systems