fix(acp): replace Goose native system prompt

[atishpatel]

Why

Buzz currently appends its managed prompt to Goose's native prompt, so managed agents receive both instruction sets instead of the intended Buzz-only system prompt.

What

• Send Goose's custom session system-prompt request with mode: "set"
• Lock the replacement contract in the ACP request test

Risk Assessment

Low — the change is limited to Goose session setup; adapters that do not implement Goose's custom method keep the existing method-not-found fallback behavior.

References

Goose v1.46.0 routes set to override_system_prompt, and its prompt builder selects that override instead of rendering the native system.md: ACP handler, prompt builder.

Validated end to end against the official Goose v1.46.0 binary with a local OpenAI-compatible capture server: the provider request contained the exact Buzz replacement prompt and did not contain Goose's native base-prompt marker.

---

Update Aug 15, 13:17 CDT: Added the Terra-high prompt-ablation comparison.

The Goose conditions used GPT 5.6 Terra at high effort on the same 11 Terminal-Bench 2.1 tasks, with two attempts per task and concurrency four. The matched append-full and set-full runs used the same persona and included the same Buzz platform prompt; Active-h is the primary measure because it excludes Buzz lifecycle overhead.

Goose condition	Pass	Active-h	Median active	Agent-h	Wall-h	Tool calls
Native prompt + Buzz prompt (append-full)	21/22	0.3042	0.85 min	0.3974	0.1496	234
Native prompt + persona only (append-persona-only)	22/22	0.3050	0.75 min	0.3990	0.1498	204
Buzz prompt replaces native prompt (set-full)	22/22	0.3340	0.87 min	0.4296	0.1551	275

Replacing instead of appending produced one additional passing attempt, but it was not an efficiency improvement in this small sample: versus append-full, set-full increased Active-h by 9.8%, median active by 2.0%, Agent-h by 8.1%, Wall-h by 3.7%, and tool calls by 17.5%. It was faster on only two of eleven per-task active-time medians (distribution-search and prove-plus-comm). With two attempts per task, these are directional results rather than confidence intervals; they support this change as an instruction-isolation/correctness fix, not a performance optimization, and argue against Goose's appended native prompt being the main source of active-time cost.

Generated with Codex

[themiguelamador]

The functional change is correct: I verified the cited Goose revision routes mode: "set" to override_system_prompt, and its prompt builder uses that override instead of the native template. The local method-not-found fallback remains intact.

The blocker is repository attribution policy:

• commit b264f5651 contains Co-authored-by: Codex <noreply@openai.com>
• the PR body ends with Generated with Codex

This repository prohibits AI attribution in both commits and PRs. An additive follow-up commit cannot remove a trailer already in history, so the commit must be rewritten/replaced and the PR footer removed.

I prepared a tree-identical signed replacement commit with no prohibited attribution: https://github.com/Complear/buzz/commit/9e509c808

Verification:

• replacement tree is byte-identical to the submitted PR head
• targeted Goose system-prompt tests: 3 passed
• cargo clippy -p buzz-acp --all-targets -- -D warnings: clean
• cargo fmt --all -- --check: clean
• git diff whitespace check passes