Menu
SkillsMP has collected 17 skills from nuncaeslupus/drawspec. Open a skill to review its source and details.
Showing 17 of 17 collected skills.
When the user wants to resume work or run the worker loop — picks the next unblocked task, optionally scoped by tag(s) and/or a workspace, or matched by title text. Use /continue [TAG … | WORKSPACE | search-text]. Do NOT use before running init.
Use whenever the user wants to turn coverage.py's coverage.json into a ranked list of the highest-value missing tests — surfaces contiguous missing-line runs marking whole untested functions or branches. A cheap line-coverage pass below mutmut-report.…
Use whenever a uv-managed Python project's dependencies need upgrading safely — the uv lock --upgrade, pip-audit, test-gate, classify-breakage runbook, splitting lockfile churn into direct vs transitive. Triggers — "upgrade the dependencies", "check for…
When the user is defining the technical solution after discovery — contracts, task split, risk register, sequencing. Owns scripts — validate_plan. Do NOT use for problem investigation (see specify), implementation (see execution), or PR review (see review).
When the user is implementing code changes from a design — code change, tests, merge-ready output. Do NOT use for investigation (see specify), design (see design), or routine one-off scripts that bypass the design step.
Use whenever the user wants an objective PASS/FAIL on a task's measurable acceptance gate from status/plan.md, with the measured numbers — reads the gate (metric, op, threshold) and recorded evidence, or audits every task's gate at once. Triggers — "did this…
Use whenever the user is creating commits, opening pull requests, or waiting on PR review/CI feedback — applies Conventional Commits + branch naming, then polls the PR for review-bot comments and CI status until it is ready to merge. Triggers — "open a PR",…
When the user needs claude-arsenal/ set up in a host repo, or wants to register a workspace via --workspace. Re-running is safe (refreshes stale bundle files only). Do NOT use to add tasks (see queue-add) or resume the worker loop (see continue).
Use when the user wants to triage a Python project's surviving mutmut mutants after a run — classifies each survivor as REAL_GAP, EQUIVALENT, or UNTESTABLE and reports the test fixes worth making. Triggers — "which mutants survived", "triage mutation testing…
Use whenever a Python package is ready to publish to PyPI — the build, twine check, version bump, tag, and upload runbook, catching version drift and stale dist/ artifacts before upload. Triggers — "publish to PyPI", "cut a release". Do NOT use to scaffold…
Use whenever a Python project needs scaffolding or retrofitting to the arsenal defaults — uv, ruff, strict mypy, the standard Makefile, a 3.12+ floor — reports the missing pieces and applies them. Triggers — "set up a Python project", "add ruff and mypy…
When the user wants to add a task to the claude-arsenal queue. Do NOT use to update or remove existing tasks.
When the user wants queue progress counts by status, or to audit the queue for inconsistencies (missing gates, missing issue handles, broken deps). Do NOT use to modify task status.
When the user is reviewing a PR, diff, design doc, or proposal — risks, tech debt, standards compliance. Do NOT use for implementation (see execution), design (see design), or release sign-off (see ship).
Use whenever the user signals end-of-job, invokes /session-end, or the opt-in SessionStart auto-fire hook runs — three steps. (1) write status/handoff.md if host repo opts in; (2) retrospective scan for repeated errors and proposed skill updates; (3) PR audit…
When the user is confirming a change is ready for production before merge — compatibility, tests, observability, rollback. Do NOT use for implementation (see execution) or PR review (see review).
When the user is investigating a problem or scoping a new feature with unclear impact — analyzes it and proposes options. Owns scripts — validate_spec. Do NOT use for already-scoped work (see design), implementation (see execution), or routine code edits.