← ClaudeAtlas

auditlisted

Sprint start: spec integrity checking, staleness scan, dependency verification, and initial CT setup. Phase /1 of the sprint pipeline. Also invoked as /1.
turqoisehex/cc-sentinel · ★ 4 · AI & Automation · score 77
Install: claude install-skill turqoisehex/cc-sentinel
# /audit — Sprint Start (alias: /1) **Trigger:** Beginning of each sprint. **Prerequisite:** Sprint N code items checked before starting N+1 (developer-parallel tasks excepted). **Channel:** CT=`CURRENT_TASK_chN.md` (channeled) or `CURRENT_TASK.md`. Scripts: `SENTINEL_CHANNEL=N`. `[chN/]`=dispatch subdir, `[_chN]`=file suffix, `[chN_]`=squad prefix. Full rules: `.claude/reference/channel-routing.md`. **Step 0:** Before any other work, TaskCreate every step. Mark in_progress→completed. ## Procedure ### Step 1: Spec integrity User specifies which spec(s). If not specified, ASK. **Full-module sweep escape hatch:** When sprint scope is full-module audit (spec > ~3000 lines OR explicit module rebuild — i.e., the sprint CT or SC entry names a single module as the rebuild target, e.g., "rebuild Auth spec", "Data-model spec rewrite", not a feature-sized edit that happens to touch the module), follow your project's spec-sweep playbook (if one exists) § Steps 1-7 in full (Steps 1-3 triage, 4-7 fix sequencing + close-out — the playbook governs the entire `/1` for full-module sweeps, not just the front end). The fidelity doctrine that governs what to verify is `~/.claude/reference/spec-verification.md` (Phases 0-5 including Phase 4.5 field-consumption audit) — same doctrine `/perfect` Phase 2.5 invokes. The playbook extends that doctrine with operational detail for full-module scale; skip the generic 6-check template in sub-step 1a below entirely when this escape hatch fires. ###