fix: address Copilot review round 12 on #2248

**The read bounds could produce a false `ok: true`.** Entries past the cap are
never fetched, `fileFailed` only examines fetched rows, and
`resource-limit-exceeded` is a WARNING — so a manifest whose 513th file was
tampered with reported the skill verified. That trades a denial of service
for a wrong answer, which is the failure this whole PR argues against. A
truncated read is now recorded as `incomplete` and forces `ok: false`,
reported separately so a consumer can tell "this skill is wrong" from "this
skill was not fully checked". And a self entry excluded by the cap is now
verified against its declared digest, not merely read for frontmatter —
otherwise the skill's own SKILL.md was the one file nobody checked.

**`entryKey` could crash the TUI.** It called `JSON.stringify` on the whole
server-controlled entry, during render. `skillEntryKey` is decided by the
same `jsonGraphError` guard that bounds the frontmatter comparison, with a
coarse fallback that still separates two unrepresentable entries by URI —
collapsing them would show one skill's verdict under another's name.

**The tab bar wraps and `tabsHeight` was hard-coded to 1.** An OAuth-capable
stdio server serving Skills needs ~107 columns, so it wraps at 132 as well as
at 80, and every pane below was sized a row too tall. `tabBarRows` derives it
from the same list `Tabs` renders, via a shared `visibleTabs(flags)` — the
duplicated filter being how the two would drift apart again.

Nine tests in a new `tabsConfig.test.ts`, including the 80-column regression.
Two assert properties rather than instances: rows never decrease as the
terminal narrows, and a tab wider than the row gets its own rather than
looping.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01ND46Ci64DC3tzBy9aG81Nc
Signed-off-by: cliffhall <cliff@futurescale.com>
This commit is contained in:
cliffhall 2026-09-07 21:57:48 -04:00
commit 27f1b2f1bd

View file

@ -513,6 +513,30 @@ export function totalSkillBytes(resources: readonly SkillResource[]): number {
);
}
/**
* A stable key for "this exact entry", safe against a hostile listing.
*
* `JSON.stringify(entry)` is the obvious implementation and is the wrong one:
* `frontmatter` is unbounded server-controlled JSON, so a deep enough object
* throws `RangeError: Maximum call stack size exceeded` — and this is evaluated
* during render, so one catalog entry could crash the pane that exists to
* report on it (Copilot).
*
* The same guard that bounds the frontmatter comparison decides it here. When
* the entry is representable the key is its serialization, which is exact;
* when it is not, the key falls back to the entry's identity plus its manifest
* length. That fallback is deliberately coarse — such an entry already carries
* a `frontmatter-unparsable` error, so what matters is that it produces a
* usable key rather than a precise one.
*/
export function skillEntryKey(entry: SkillEntry): string {
if (jsonGraphError(entry) !== undefined) {
const count = Array.isArray(entry.resources) ? entry.resources.length : -1;
return `${skillUriIdentity(entry.uri)}#unrepresentable:${count}`;
}
return JSON.stringify(entry);
}
/**
* Whether a `skills/get` entry describes the same skill as the `skills/list`
* entry alongside it, compared **semantically** rather than byte-for-byte.