fix(buzz-acp): loosen workspace-scan guardrail to allow named paths

[salman1993]

Problem

Agents refused to read a specific, user-named path outside .buzz (e.g. ~/.claude/skills/context-health-check/SKILL.md), and behaved inconsistently — refusing in one channel, listing the same dir in another. This is the issue Aaron and Wes hit: buzz://message?channel=e62570dd-33ad-42c5-b92b-75f2689f9694&id=401199157f8a9734109fdddfbbce4603a1ff465fae895042f29738ad52a5d15f

Root cause: two prompt fragments told agents not to search $HOME "or other directories". The intent (added in #1194 by Will Pfleger) was narrow — stop unscoped $HOME sweeps that fire macOS TCC dialogs or make a model invent its own workspace dir. But "or other directories" reads as a blanket filesystem wall, so a cautious model generalized it to "any path outside .buzz" and refused a targeted, named read.

Fix

Prompt-only. Reframe both fragments per context-engineering guidance — state the reason and trust judgement, rather than banning a technique:

crates/buzz-acp/src/pool.rs — workspace_section()

…This is where you already are, so start here rather than scanning $HOME. Any specific path the user names is fine to read.

crates/buzz-acp/src/base_prompt.md line 106

These paths are relative to your working directory — start there for your own files rather than scanning $HOME or /. When the user names a specific path, read it.

Preserves the original anchor intent (start at the grounded workspace; avoid unscoped $HOME sweeps) while explicitly permitting a named-path read — the exact case that was broken.

Testing

cargo test -p buzz-acp — 816 unit + 9 integration tests pass, 0 failures. Prompt-only change; the doc comment at pool.rs:1327-1333 already states the narrow intent and is left as-is.

[salman1993]

🤖 Benchmark result using read-named-path-outside-workspace from the Buzz-native dataset in #6264.

Test setup:

• Combined this PR's prompt commit (37df12ce) with the dataset from Add Buzz-native collaboration benchmarks #6264
• Claude Sonnet 4.6
• 5 concurrent attempts

Results:

• PR fix(buzz-acp): loosen workspace-scan guardrail to allow named paths #6261 prompt: 5/5 passed with reward 1.0; zero exceptions, refusals, or missing final responses
• Baseline prompt without fix(buzz-acp): loosen workspace-scan guardrail to allow named paths #6261: 3/5 passed; one explicit workspace-boundary refusal and one missing final response
• Every scored component for the fix(buzz-acp): loosen workspace-scan guardrail to allow named paths #6261 runs was 1.0: named-path read, action reported, refusal-free, evidence complete, expected author, and same channel

One separate tool ergonomics issue remains: all five runs first passed literal ~ to read_file, which resolved incorrectly as /app/~. Every agent recovered and successfully read the named external file, so this did not affect the benchmark result.