skills/sentry-triage-to-pr/SKILL.md
Triage unresolved Sentry issues into ranked groups, GitHub issue plans, branches, subspawn worktree assignments, PRs, and closeout loops using the sentry CLI, GitHub CLI, and local verification. Use when asked to prioritize Sentry backlogs, group production issues, create GitHub issues or PRs from Sentry evidence, or parallelize Sentry fixes.
npx skillsauth add bjornmelin/dev-skills sentry-triage-to-prInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this skill as the front door for turning unresolved Sentry issues into
reviewable, verified fixes. Keep sentry-cli-fix-issues focused on one issue;
this skill owns backlog ranking, grouping, GitHub issue planning, branch/PR
handoff, and PR closeout coordination.
sentry CLI first. Use sentry api only for gaps such as advanced
sorts, external links, tag values, and dry-run write previews..codex/sentry/<timestamp>-<slug>/ in the
target repo. Do not write generated bundles into the skill directory.--apply command outside the operator.sentry issue explain and sentry issue plan as
advisory evidence only. Confirm with stack frames, repository code, and tests.Run the portable operator from the target repository to preflight local tools,
capture redacted Sentry evidence, rank unresolved issues, group them into fix
units, render GitHub issue plans, and plan branch/worktree assignments. Use
references/workflow.md for the detailed command sequence and closeout rules.
scripts/sentry_triage_operator.py: portable read-only operator for capture,
scoring, grouping, GitHub plans, worktree plans, and bundle validation.references/workflow.md: full preflight, capture, grouping, GitHub planning,
worktree planning, implementation, and post-ship Sentry workflow.references/operator-contract.md: schema, scoring, redaction, and generated
artifact details.references/github-and-pr-closeout.md: GitHub issue, branch, PR, and babysit
rules.development
Pre-PR multi-model review, parallel opus and codex exec adversarial lanes, then adversarial verification of merged findings. Read-only. Use before shipping nontrivial diffs.
tools
Independent gpt-5.6 diff review via the Codex CLI, normal or steerable adversarial with JSON findings. Use before shipping nontrivial changes.
development
Delegate implementation, investigation, or bulk work to gpt-5.6 codex via pinned codex exec. Use for clear-spec builds, migrations, debugging, or any task MODELS.md routes to codex.
development
Adversarial pre-mortem: imagine the plan failed, work backwards to surface risky assumptions + irreversible bets, then harden them. Proactively offer it (after the current request; confirm first) before a hard-to-reverse or one-way-door call (API, schema, framework, a hire), an all-upside plan, or unvalidated assumptions. Also on request.