refactor(prompt): simplify Buzz agent guidance

[salman1993]

Why

The Buzz agent prompt carries verbose agent-draft instructions that add prompt weight without improving benchmark correctness. Buzz-native channel trials also showed agents creating permanent channels while claiming they had applied a one-hour TTL. The static CLI inventory remains for efficient command discovery by smaller agents.

What

• Retain the static Buzz CLI command table and direct agents to use --help for full usage
• Compact the owner-reviewed agent-draft guidance without changing its workflow
• Clarify that buzz channels create is permanent unless --ttl is provided

Benchmark Results

• Terminal-Bench retained 22/22 correctness in the initial no-table candidate while reducing prompt size, input tokens, tool calls, and cost; total active time was effectively flat.
• Across two Buzz-native runs, the non-channel cases tied at 26/30; the CLI clarification directly addresses the observed TTL omission.

Update — 2026-08-20

• Restored the CLI command table following review feedback about multistep discovery with smaller agents. The restored-table candidate has not been re-benchmarked.

[philazar]

lgtm

[atishpatel]

Doesn't removing this mean the agent will always have to make multiple tool calls to see available buzz cli commands?

Discovery with smaller agents will become a real issue if it's multistep

[atishpatel]

Maybe need to pair this with buzz --help output improvement?
In the help output you can include subcommands so it's not multiple calls

[salman1993]

i think for most turns, the agent doesn't need to call buzz CLI commands outside of buzz messages which is already mentioned in the prompt in the section below. Its better to progressively disclose. In the buzz-native benchmark, i didn't see much of a change when this table was removed.

Run buzz --help or buzz <group> --help for full usage. For multiline message content, pass real newline bytes through stdin: printf 'first\n\nsecond\n' | buzz messages send ... --content -. Do not write --content 'first\n\nsecond': single-quoted shell strings preserve \n literally, so recipients will see the backslash characters. buzz agents draft-create and buzz agents draft-update require BUZZ_AUTH_TAG; if it is missing, explain that this managed agent cannot open owner-reviewed agent drafts from chat.

[atishpatel]

Improve the --help output so subcommands are visible