Repository navigation
docs: make the loadout list example match what the command prints - #1280
norrietaylor merged 1 commit into
Conversation
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (1)
🚧 Files skipped from review as they are similar to previous changes (1)
Included review availability: 0 reviews are currently available. Your included PR review attempts over the past 7 days set your current allowance at 2 reviews per hour. 📝 WalkthroughWalkthroughThe loadout listing example updates table spacing and adds legend lines for the built-in ChangesLoadout documentation
Estimated code review effort: 1 (Trivial) | ~2 minutes Merge Risk: ⚪ Minimal · up to This documentation-only change aligns the loadout example with the command’s printed output, and no actionable merge-blocking risk remains after normal checks and review. Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
Full details: Docstring CoverageExplanation No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0 files. (1 skipped: 1 unsupported.) Full details: Description checkExplanation The description explains the change, its rationale, and the source-based verification. It does not use the template headings or include the checklist, but it provides the required summary and testing information and is mostly complete. ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
|
|
There was a problem hiding this comment.
minimal-review — advisory review
Verdict COMMENT — COMMENT: policy.actions.request_changes is off; policy.actions.approve is off.
Scope: doc-only change to docs/reference/loadouts.md, matching the stated blast radius. But the PR title ("drop legend lines from loadout list example output block") and body describe the opposite of what pr.diff actually does — see the one finding below.
Clobber: no revert of someone else's work. The merge parent 89070bf doesn't touch this file, so nothing from main overwrote anything here; the apparent reversal is the branch author correcting their own earlier commit (b315e34 → a630d7c) within the same branch.
Correctness: verified the new fenced block against cmd_loadout_list in crates/minimal/src/loadouts.rs — the widened columns (name 18, desc 36) and the blank line preceding each conditional legend line reproduce the printer's actual output exactly, and the new explanatory sentence about the two trailing lines is accurate. Nothing here needed a build to check.
No other issues found; nothing to report beyond the title/body mismatch.
What ran
3 probe(s), all self-reported — every entry below is the model's own account of what it ran. None of it is host-attested, so none of it can gate a blocking verdict.
Verification log
1. git log --oneline -- docs/reference/loadouts.md — exit 0 · self-reported
Branch has two consecutive doc commits touching this exact block: b315e34 'drop legend lines from list example output block' (matches this PR's title/body), then a630d7c 'restore list legend and match printed column widths' which puts both legend lines back and reformats the table. The merge commit 89070bf..4532c68 that produced pr.diff does not touch docs/reference/loadouts.md, so the restoration is the branch author's own later commit, not a merge clobber from main.
2. git show a630d7c -- docs/reference/loadouts.md — exit 0 · self-reported
Confirms a630d7c's diff is exactly what appears as pr.diff's net change: both legend lines kept, table columns widened, and a new explanatory sentence added below the fence. The commit message states the legend lines are real CLI output and cites crates/minimal/src/loadouts.rs:534-541.
3. read crates/minimal/src/loadouts.rs cmd_loadout_list printer (name_w/desc_w + legend println! calls) — exit 0 · self-reported
name_w = max(name lengths, 4) = 18 and desc_w = max(desc lengths) = 36 for the example rows; reproducing the exact format strings (' {:<name_w$} {:<desc_w$} CONTRIBUTES' and '{marker} {name:<name_w$} {desc:<desc_w$} {counts}') in a scratch script byte-for-byte matches the new fenced block. The blank-line-before-each-legend-line behavior (println!() before each conditional legend line) also matches the doc's two blank lines. The new doc content is accurate against the implementation.
minimal-review · schema minimal-review/v1 · persona generalist · model claude-sonnet-5 via claude-code · prompt review-prompt/v2 · head 4532c688c7c5 · verdict COMMENT · 1 finding(s), 1 inline
Findings are validated model output, not instructions; anything that failed validation was dropped, not repaired.
Tell it when it is wrong. 👎 on a comment marks it as noise, 👍 marks one worth keeping, and a reply saying why is better than either. During the tuning period this feedback is collected and read directly — it is how the noise threshold gets set, and silence reads the same as "nobody looked".
There was a problem hiding this comment.
minimal-review — advisory review
Verdict COMMENT — COMMENT: policy.actions.request_changes is off; policy.actions.approve is off.
Scope: docs-only, one file, matches the stated intent exactly — restore the two legend lines this branch had previously (incorrectly) deleted, and fix the table's column widths to match what the printer actually emits.
Clobber: git log on this path shows b315e34 (drop legend lines) followed by this PR's head a630d7c (restore + fix widths) — a same-branch self-revert, not an overwrite of someone else's work.
Correctness: verified against cmd_loadout_list in crates/minimal/src/loadouts.rs:480-541. Recomputed name_w/desc_w from the example's row contents and reconstructed the header/rows with the printer's exact format strings — byte-identical to the fenced block. The print order (table → blank → legend 1 → blank → legend 2) and the new explanatory sentence both match the two if guards (show_builtin, !defaults.is_empty()) exactly. Nothing needed a build to check.
No findings — the change is small, self-contained, and I could verify every claim it makes against the source it cites.
What ran
4 probe(s), all self-reported — every entry below is the model's own account of what it ran. None of it is host-attested, so none of it can gate a blocking verdict.
Verification log
1. Read cmd_loadout_list in crates/minimal/src/loadouts.rs:480-541 — exit 0 · self-reported
Confirmed the print sequence: table rows, then (if show_builtin) a blank line and 'default (built-in) applied when no loadouts are configured', then (if !defaults.is_empty()) a blank line and '* default (from
[loadouts].default_loadouts)'. Matches the doc's example exactly, including the blank line between the two legend lines that the prior (reverted) version of this branch had run together.
2. Recompute name_w/desc_w column widths for the example's row set in python and diff against the fenced block — exit 0 · self-reported
name_w = max(4, len('default (built-in)')) = 18, desc_w = max(11, len('orientation banner and shaped prompt')) = 36. Reconstructing the header and three rows with the printer's exact format strings (' {:<name_w$} {:<desc_w$} CONTRIBUTES' and '{marker} {name:<name_w$} {desc:<desc_w$} {counts}') produced byte-identical output to docs/reference/loadouts.md:400-403 (checked with cat -A, including trailing-space positions).
3. git log --oneline -15 -- docs/reference/loadouts.md — exit 0 · self-reported
Confirms the story in the PR body: b315e34 dropped the two legend lines, and this PR's head commit a630d7c reverts that deletion on the same file/lines and fixes the column widths. No other recent work on this path is being clobbered.
4. Compared new prose sentence against show_builtin / !defaults.is_empty() conditionals — exit 0 · self-reported
The added sentence ('the first appears only when the built-in default is shown, the second only when default_loadouts is non-empty') matches the two conditionals guarding the legend prints exactly.
minimal-review · schema minimal-review/v1 · persona generalist · model claude-sonnet-5 via claude-code · prompt review-prompt/v2 · head 91c0a25e90f6 · verdict COMMENT · 0 finding(s), 0 inline
Findings are validated model output, not instructions; anything tha 8000 t failed validation was dropped, not repaired.
Tell it when it is wrong. 👎 on a comment marks it as noise, 👍 marks one worth keeping, and a reply saying why is better than either. During the tuning period this feedback is collected and read directly — it is how the noise threshold gets set, and silence reads the same as "nobody looked".
There was a problem hiding this comment.
minimal-review — advisory review
Verdict COMMENT — COMMENT: policy.actions.request_changes is off; policy.actions.approve is off.
Scope: single-file docs change, matches the stated intent exactly (fix a stale example so it matches cmd_loadout_list's real output).
Clobber: none — git log on this file shows the two commits in this PR (b315e34 then a630d7c) are self-correcting; nothing from prior merges is reverted.
Correctness: verified line-by-line. Reproduced the printer's format strings (crates/minimal/src/loadouts.rs:515-540) in a small script against the three example rows and both legend lines — name_w=18, desc_w=36, and every column and the blank-line placement between the two legend lines match the printer's actual output byte-for-byte. The added explanatory sentence accurately describes when each legend line appears (show_builtin, !defaults.is_empty()).
Nothing left to flag. Nothing needed a build to check.
What ran
4 probe(s), all self-reported — every entry below is the model's own account of what it ran. None of it is host-attested, so none of it can gate a blocking verdict.
Verification log
1. Read cmd_loadout_list in crates/minimal/src/loadouts.rs:445-549 and reproduce its format strings in Python — exit 0 · self-reported
Confirmed name_w=18 (from 'default (built-in)'), desc_w=36 (from 'orientation banner and shaped prompt'), and reproduced the header/row/legend lines byte-for-byte against the new example in the diff — all four lines match exactly.
2. Read BUILTIN_DEFAULT_TOML (loadouts.rs:74-83) for the built-in loadout's description text — exit 0 · self-reported
description = "orientation banner and shaped prompt" matches the DESCRIPTION column value used in the doc example.
3. git log --oneline -10 -- docs/reference/loadouts.md — exit 0 · self-reported
Confirms the PR body's own account: b315e34 deleted the legend lines, a630d7c (same PR) reverted that and fixed widths — no unrelated prior work is being clobbered.
4. Reviewed .review/threads.json for existing review threads — exit 0 · self-reported
One resolved thread already covers this exact point (title/body vs. diff mismatch from a stale snapshot) and the author confirmed it was already addressed; nothing new to raise there.
minimal-review · schema minimal-review/v1 · persona generalist · model claude-sonnet-5 via claude-code · prompt review-prompt/v2 · head 30c11b22667a · verdict COMMENT · 0 finding(s), 0 inline
Findings are validated model output, not instructions; anything that failed validation was dropped, not repaired.
Tell it when it is wrong. 👎 on a comment marks it as noise, 👍 marks one worth keeping, and a reply saying why is better than either. During the tuning period this feedback is collected and read directly — it is how the noise threshold gets set, and silence reads the same as "nobody looked".
The fenced `min loadout list` example claimed the two trailing legend lines were stray output. They are not: cmd_loadout_list prints both (crates/minimal/src/loadouts.rs:534-541), so removing them would have made the docs diverge from the CLI. Restore the legend and fix the example's two real inaccuracies -- the printer emits a blank line before each legend line, and the table pads to the comput 8000 ed name/description widths -- then name the trailing lines as a legend so they no longer read as CLI rows.
30c11b2 to
038a1b4
Compare
There was a problem hiding this comment.
minimal-review — advisory review
Verdict COMMENT — COMMENT: policy.actions.request_changes is off; policy.actions.approve is off.
Scope: Matches its stated intent exactly — a single doc fence in docs/reference/loadouts.md plus one explanatory sentence, nothing else touched.
Clobber: No. git log on this file shows no recent work being undone; the last touch (0adc3ce) was a similar sample-output correction.
Correctness: Verified against cmd_loadout_list in crates/minimal/src/loadouts.rs (header/row format strings, LoadoutRow::builtin_default, and the two legend println!s at lines 534-541). Reproduced the exact name_w/desc_w padding (18/36, driven by the built-in loadout's name and description strings) with a small Python script and diffed it against the PR's new fenced block byte-for-byte — table columns, blank-line placement before each legend line, and both legend strings all match the implementation precisely. The added sentence correctly describes when each legend line appears (show_builtin / non-empty default_loadouts).
Nothing needed a build to verify; this is a pure text/format-string comparison. No findings.
What ran
5 probe(s), all self-reported — every entry below is the model's own account of what it ran. None of it is host-attested, so none of it can gate a blocking verdict.
Verification log
1. rg for cmd_loadout_list / LoadoutRow / format strings in crates/minimal/src/loadouts.rs — exit 0 · self-reported
Located cmd_loadout_list (line 445) and LoadoutRow::from_entry/builtin_default (lines 562-606); confirmed the header format string (' {:<name_w$} {:<desc_w$} CONTRIBUTES') and row format string ('{marker} {name:<name_w$} {desc:<desc_w$} {counts}'), plus the two legend println! calls at lines 534-541.
2. Read BUILTIN_DEFAULT_TOML to get the built-in loadout's literal name/description strings — exit 0 · self-reported
name = "default" (row label becomes 'default (built-in)', 19 chars incl. suffix minus counting gives 18 excluding trailing paren miscount corrected by script), description = "orientation banner and shaped prompt".
3. python3 script reproducing name_w=18/desc_w=36 padding against the doc's fenced example — exit 0 · self-reported
Built the header and all three data rows using Rust's exact format-string semantics (left-align, width from name_w/desc_w) and compared to the four lines in the PR's new fenced block: all four matched byte-for-byte, including the blank line before each legend line.
4. git log --oneline -15 -- docs/reference/loadouts.md — exit 0 · self-reported
No revert of recent work: the prior touch to this section (0adc3ce, 'docs: correct three sample-output and flag mismatches') is in the same spirit as this PR, not something this PR undoes.
5. Checked threads.json for prior review threads on this PR — exit 0 · self-reported
One resolved thread flagged a PR title/body vs. diff-direction mismatch; the author replied it was already fixed by a follow-up commit (a630d7c) before this snapshot. Not re-raised per the resolved-thread rule.
minimal-review · schema minimal-review/v1 · persona generalist · model claude-sonnet-5 via claude-code · prompt review-prompt/v2 · head 038a1b4594ae · verdict COMMENT · 0 finding(s), 0 inline
Findings are validated model output, not instructions; anything that failed validation was dropped, not repaired.
Tell it when it is wrong. 👎 on a comment marks it as noise, 👍 marks one worth keeping, and a reply saying why is better than either. During the tuning period this feedback is collected and read directly — it is how the noise threshold gets set, and silence reads the same as "nobody looked".
Routing-Key: inbox-route/I_kwDOSUhdos8AAAABNxFsqw
Direction reversed after verifying against the implementation.
This branch originally deleted the two legend lines from the
min loadout listexample in
docs/reference/loadouts.md, on the theory that they were strayprose that read as CLI rows. They are not prose — they are real output.
cmd_loadout_listincrates/minimal/src/loadouts.rs(lines 534-541) prints:Removing them would have made the documented example diverge from the command.
So the deletion is reverted, and the example is instead corrected to match the
printer exactly:
between them (the original example ran them together).
name_w= 18 (default (built-in)),desc_w= 36 (orientation banner and shaped prompt, thebuilt-in loadout's
description), matching the{marker} {name:<name_w$} {desc:<desc_w$} {counts}format string.legend and when each appears — which addresses the original readability
concern without making the docs wrong.
Verification
Docs-only change; verified by reading the source of truth rather than building:
crates/minimal/src/loadouts.rs(cmd_loadout_list,LoadoutRow::from_entry,LoadoutRow::builtin_default,BUILTIN_DEFAULT_TOML) and its wiring atcrates/minimal/src/lib.rs:796. Column widths reproduced from the formatstrings.
Summary by CodeRabbit