Sprint scaffold
Scaffold a new Aria sprint folder with Filing Cabinet spec_charter.md + Flowstate .claude/agents/pr-review.md + reference/sql/ subdirs. Use when starting a new sprint (e.g. "open sprint 021", "start a sprint for X", "new sprint folder"). Drops a working skeleton in /opt/aria/v4/sprints/NNN_slug/ with all the load-bearing files pre-filled so every shipped item gets an audit-trail home. Validated in Sprint 020 (two-terminal push, 20+ items shipped under this pattern).From its SKILL.md
npx -y skills add iansteitz1-eng/aria-skills --skill sprint-scaffoldAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
6.2 KB, ~1.6k tokens by cl100k_base, as published. Nobody here has run it
Sprint Scaffold
Apply the Filing Cabinet ($49 product) spec_charter.md template + the Flowstate ($299 product) .claude/agents/pr-review.md template to internal Aria sprints (not just customer-facing project tarballs). Closes the audit-trail loop: every sprint has a mission, scope, risks, acceptance criteria, and a review rubric for everything shipped under it.
This skill emerged from Sprint 020 (two_terminal_push, 2026-05-23) where T1 and T2 shipped 11 items in parallel with zero conflicts because the sprint folder enforced clean ownership boundaries up front.
When to use
- User says "open sprint NNN" / "start sprint for X" / "new sprint" / "scaffold a sprint" / "let's sprint on Y"
- You're about to start work that spans 3+ items + multiple sessions and should have one canonical home
- Multiple terminals or contributors need a file-lock contract (this skill ships HANDOFF_*.md templates)
Steps
-
Pick the sprint number. Look at
/opt/aria/v4/sprints/— find the highest existing NNN and add 1. Slug is lowercase-hyphenated.ls /opt/aria/v4/sprints/ | sort -n | tail -3 -
Create the folder + subdirs:
SPRINT=021_<slug> mkdir -p /opt/aria/v4/sprints/$SPRINT/{reference,sql,.claude/agents} -
Write spec_charter.md (Filing Cabinet pattern). Use the template below — pre-fill what you know, leave the rest as
TODO:# Sprint NNN — <Title> · spec_charter v1 **Opened:** YYYY-MM-DD **Closes:** YYYY-MM-DD (target) **Method:** Filing Cabinet spec_charter pattern + Flowstate pr-review pattern ## 1. Mission <1-2 sentences. What changes if this sprint succeeds?> ## 2. Stakeholders | Role | Person | Responsibility | |---|---|---| | Owner | Ian | Final call | | T1 | Claude | <lane> | | T2 | (if applicable) | <lane> | | Informed | Stephen | <when needed> | ## 3. Constraints 1. <calendar / doctrine / file-lock / restart / etc.> ## 4. Scope ### P0 (must ship) | # | Item | Owner | Status | ### P1 (close if possible) ### P2 (stretch) ### Out of scope (defensibly deferable) ## 5. Risks | Risk | Likelihood | Mitigation | ## 6. Open questions ## 7. v1 acceptance criteria - [ ] <criterion> ## 8. Versioning Per Filing Cabinet versioning protocol: copy to spec_charter_v2.md when scope changes. Never edit v1 in place. ## 9. Tools/skills used (reuse log) | Tool/skill | Used for | New or existing | ## 10. New skill emerging (if any) -
Write .claude/agents/pr-review.md (Flowstate pattern). Use this 5-dim rubric template:
# Sprint NNN PR-Review Rubric (Flowstate pattern, sprint-local) Apply to every item shipped before marking done. ## The 5 dimensions ### 1. Correctness — does it do what the charter §4 said? ### 2. Security — auth gates? secrets? injection? ### 3. Test coverage — smoke trace saved? ### 4. Doctrine alignment — which feedback_* rules apply? ### 5. Follow-ups — what was deferred? ## Output formatSprint NNN self-review — <item slug>
1. Correctness: ✅ / ⚠️ / ❌ — <one-line evidence> 2. Security: ✅ / ⚠️ / ❌ 3. Test coverage: ✅ / ⚠️ / ❌ 4. Doctrine: ✅ / ⚠️ / ❌ 5. Follow-ups: <ops_log ids, skill ideas> Verdict: APPROVE / REQUEST CHANGES / RETHINK Smoke command: <copy-paste-able>
-
Drop a HANDOFF template (if multi-terminal):
touch /opt/aria/v4/sprints/$SPRINT/HANDOFF_T1_TO_T2.mdPre-fill with the file-lock contract + route-prefix partition pattern from the Sprint 020 handoff.
-
Insert a parent row in v4_ops_log:
INSERT INTO v4_ops_log (priority, status, category, area, title, detail, owner, purpose) VALUES ('P0','in_progress','product','aria_code', 'Sprint NNN — <title>', 'Charter at /opt/aria/v4/sprints/NNN_slug/spec_charter.md', 'claude','sprint_NNN');Subsequent items in this sprint tag
purpose='sprint_NNN'for clean filtering. -
Report. Print the absolute path to the new sprint folder + the next 1-3 items to populate.
Folder layout this skill creates
/opt/aria/v4/sprints/NNN_slug/
├── spec_charter.md ← Filing Cabinet pattern
├── .claude/agents/pr-review.md ← Flowstate pattern
├── HANDOFF_T1_TO_T2.md ← Inter-terminal coordination (template)
├── reference/ ← Source docs that fed the sprint
└── sql/ ← Schema migrations this sprint introduces
Notes
- This skill was promoted from Sprint 020 (
two_terminal_push, 2026-05-23) where the pattern proved itself shipping 11 items in parallel between T1 and T2 with zero file collisions. Seeproject_sprint_020_t1a_60s_warning_shipped(and 6 sibling memories) for examples of the pattern in use. - The
spec_charter.mdtemplate here is the same one that ships at the root of every Filing Cabinet customer tarball, just retargeted at our internal sprint context. The Mission/Stakeholders/Constraints/Risks/Acceptance pattern works identically for both. - The
pr-review.mdtemplate here matches the Flowstate scaffold pack —.claude/agents/pr-review.mdat sprint root means Claude Code agents picked up later in the sprint will auto-honor the rubric. - Charter versioning rule (never edit v1 in place; copy to v2 if scope changes) is load-bearing. Audit trail needs the WHY of pivots.
- Pair with the existing
session-endskill — at sprint close, the per-item memories that piled up should each have a## Sprint NNN self-reviewblock already. - Cross-link sprint memories with
[[other-slug]]for navigability.
What ships with it: 9 files
14.4 KB alongside SKILL.md, 1 of them executable
manifest/
- gemini.json1.3 KB
- mcp.json1.2 KB
- openai.json1.2 KB
- README.md1.1 KB
templates/
- HANDOFF_T1_TO_T2.md1.5 KB
- pr-review.md2.5 KB
- spec_charter.md2.2 KB
- README.md1.8 KB
- sprint_scaffold.shruns1.6 KB