fix(cli): bound large gh output

This commit is contained in:
Codex
2026-05-23 10:48:24 +00:00
parent 152ded3a7b
commit 431d6bd25b
4 changed files with 158 additions and 4 deletions
+1
View File
@@ -5575,6 +5575,7 @@ export function ghHelp(): unknown {
"PR list defaults to --state all for compatibility with earlier UniDesk CLI behavior; supported states are open, closed, and all.",
"issue read is the canonical read path; view remains a compatibility alias. Read/view accept owner/repo#number shorthand and derive --repo unless an explicit conflicting --repo is supplied, which fails structurally with suggested commands. Read supports legacy --json field selection such as --json body and still exposes .data.issue.body for compatibility; unsupported fields fail structurally.",
"--raw and --full are explicit full-disclosure aliases for gh issue read/view and gh pr read/view. They request the full supported read/view JSON field set while keeping default command output structured JSON.",
"GitHub CLI output larger than 20 KiB is automatically written to /tmp/unidesk-cli-output/*.json; stdout stays bounded JSON with outputTruncated=true, the dump path, total bytes/lines, and head/tail previews.",
"issue create accepts repeatable --label values and comma-separated labels; dry-run prints the parsed labels and non-dry-run sends them in the GitHub REST create-issue payload.",
"--body-file is the recommended source for Markdown bodies so real newlines, backticks, and tables are read as file bytes instead of shell arguments.",
"update defaults to --mode replace; --mode append reads the current body and appends file bytes so real newlines, backticks, and Markdown tables are preserved.",