← ClaudeAtlas

falisted

Apply every finding from the most recent review, audit, verification, or check in this session
svyatov/agent-toolkit · ★ 0 · AI & Automation · score 76
Install: claude install-skill svyatov/agent-toolkit
Apply the findings from the most recent review, audit, verification, lint run, test run, or other check reported earlier in this session, whatever produced it. $ARGUMENTS narrows which ones; empty means all of them, at every severity. A finding the report left as a choice for the user: apply the option $ARGUMENTS names, else the one the report recommended in so many words ("I recommend X"), worded exactly as it was offered. An option listed first or labelled "proposed" is not a recommendation. With none named, skip the finding as waiting on that choice. 1. No such report in this session: say so and stop. Do not go looking for problems to invent. 2. Fix root causes, not the symptom each finding names. Grep every caller before editing a shared function: one guard where the callers converge beats a guard in each of them. 3. Do not expand scope. A finding is a fix, not an invitation to refactor around it. When the finding proposes a fix, apply that fix, together with any change it cannot work without, and name that change on the finding's report line. If it cannot work even so, skip the finding and say why instead of choosing another. 4. For every repository you edited, this one or another, read its `AGENTS.md`, `CLAUDE.md` and `CONTRIBUTING.md`. Run the checks they name, and make the other edits they require of a change (a version bump, a changelog entry). Done when each edited repository has its checks run, or is named in the report as having none after yo