Answer: Read the check run's `output.text` field and parse the last line, which is an HTML comment holding the severity counts as JSON, using `gh` and jq. The command scans for the pattern `^$` and converts it with `fromjson`. Status: Confirmed Entry: tool/claude-code/integrations/read-review-result-from-script Revision: 2 Digest: sha256:4ddac2aaf08095ad (changes whenever the content changes) Timeline: Created: 2026-11-13T03:03:44Z Last changed: 2026-11-19T12:15:10Z (scan pattern: — → ^$) Last verified: 2026-11-20T12:19:27Z Next check: 2026-11-21T12:19:27Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of field to read | output.text | [1] | 2026-11-20 severity line format | | [1] | 2026-11-20 scan pattern | ^$ | [1] | 2026-11-20 example output | {"nit": 1, "normal": 2, "pre_existing": 0} | [1] | 2026-11-20 missing line error | no severity line | [1] | 2026-11-20 Sources: [1] Anthropic. "Read and request reviews on a pull request". https://code.claude.com/docs/en/code-review-pull-requests.md Role: primary. Accessed: 2026-11-20T12:19:27Z Quote: "Read the counts from the check run's `output.text` field, which holds the Details text." (Supports: field to read) Quote: "" (Supports: severity line format) Quote: "scan("^$")[0] | fromjson" (Supports: scan pattern) Quote: "for example `{"nit": 1, "normal": 2, "pre_existing": 0}`" (Supports: example output) Quote: "the command prints an error that contains `no severity line` instead of counts" (Supports: missing line error) Published: 2026-11-20T03:42:08Z Snapshot SHA-256: 10443e3e651b5ae3d7dcc4bbc3ccb687de4848ead0084510be5bc37d7e4a32a7 (captured 2026-11-19T12:15:09Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-19T12:15:10Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T12:19:27Z. No independent second source is configured for this entry Last successful check: 2026-11-20T12:19:27Z Last attempt: 2026-11-20T12:19:27Z Notes and limitations: Extracted automatically from a single official source (section "Read the result from a script"). Related: https://latestn.com/tool/claude-code/integrations/request-review-on-pull-request How do I request a Claude Code review on a pull request? https://latestn.com/tool/claude-code/integrations/respond-to-review-findings How do I respond to or dismiss a Claude Code review finding on my pull request? https://latestn.com/tool/claude-code/integrations/resume-session-typescript-sdk How do I resume a saved session in Claude Code with the TypeScript Agent SDK? https://latestn.com/tool/claude-code/integrations/plugin-updates-renames-removals How do I ship updates, renames, or removals of a Claude Code plugin to users? https://latestn.com/tool/claude-code/integrations/project-github-access How do I set up GitHub access for a Claude Code project? https://latestn.com/tool/claude-code/integrations/publish-plugin-own-marketplace How do I publish and distribute a Claude Code plugin through my own marketplace? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-19T12:15:10Z | changed | scan pattern: — → ^$ | [1] 2026-11-13T03:03:44Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/claude-code/integrations/read-review-result-from-script/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/claude-code/integrations/read-review-result-from-script POST: https://latestn.com/dispute/tool/claude-code/integrations/read-review-result-from-script with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/claude-code/integrations/read-review-result-from-script?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). How do I read Claude Code review results from a script using the check run output? Revision 2, last verified 2026-11-20T12:19:27Z. https://latestn.com/tool/claude-code/integrations/read-review-result-from-script (accessed 2026-11-21T09:59:19Z). @misc{LatestN_tool_claude_code_integrations_read_review_result_from_script_r2, title = {How do I read Claude Code review results from a script using the check run output?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/claude-code/integrations/read-review-result-from-script}}, note = {Revision 2; digest sha256:4ddac2aaf08095ad; last verified 2026-11-20T12:19:27Z; accessed 2026-11-21T09:59:19Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/claude-code/integrations/read-review-result-from-script