Repository navigation
feat(issue-import-from-scan): attach a scan's candidate patches to new trackers - #1896
Merged
Merged
Conversation
…w trackers When a scan ships candidate patches, import-from-scan now puts them on each tracker it creates, so whoever picks up the fix starts from them instead of digging through the decrypted archive. - attach_archive lists each finding's patches/<n>.patch files, in numeric order, as `patches` (`[]` when none). - New `render-patch-comment` command writes the tracker comment: a short untested-and-private warning, then one collapsed <details> block per candidate, titled by the patch's own header line and fenced as diff with a fence longer than any backtick run inside it. @-tokens in the titles are code-wrapped so they notify nobody. Bodies stay under GitHub's comment limit by splitting across several comments; a single patch too large for one comment is refused rather than cut. Each body starts with the `scan-patches` skill-comment marker, so sync's pre-flight reads it as skill-authored, not as human activity. - Step F posts the comment after the operator confirms; the patches are quoted only, never applied or run, and never copied into a public PR. - The scan-format contract gains the optional `patches` field. Generated-by: Claude Opus 5 Claude-Session: https://claude.ai/code/session_01NyQq9UPScpux59zQR38WWD
…the patch comment The scan's target and commit come from the disclosure email. A value holding a backtick would close the code span and let the email inject Markdown or an @-mention into the tracker comment, so a value that is not a plain ref is left out of the intro. Generated-by: Claude Opus 5 Claude-Session: https://claude.ai/code/session_01NyQq9UPScpux59zQR38WWD
potiuk
enabled auto-merge (squash)
October 29, 2026 09:24
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Some scans ship candidate fixes alongside their findings, for example the
patches/<n>.patchfiles in a decrypted findings archive. Until nowsecurity-issue-import-from-scanlinked only each finding's write-up. Whoever picked up a new tracker had to go back into the archive to find the patches.This change puts them on the tracker:
attach_archivelists each finding's candidate patches in numeric order aspatches([]when none). It only lists them, never runs anything.render-patch-commentcommand writes the tracker comment. It starts with a short warning that the patches are untested and private, then has one collapsed<details>block per candidate. Each block is titled by the patch's own header line and fenced asdiff, with a fence longer than any backtick run inside the patch.@tokens in the titles are wrapped in backticks so they notify nobody.patchesfield.Tests cover ordering, folded rendering, fence escaping, splitting, the oversize error, and the refusal when no archive was linked.
🤖 Generated with Claude Code
https://claude.ai/code/session_01NyQq9UPScpux59zQR38WWD