Sprint

Sprint Skill — Generic Sprint Management for bkit Users

Sprint = meta-container above bkit's PDCA 9-phase. A sprint groups one or more features under a shared scope, budget, and timeline. Each sprint runs its own 8-phase lifecycle: prd -> plan -> design -> do -> iterate -> qa -> report -> archived.

Quick Start

/sprint init my-launch --name "Q2 Launch" --trust L3
/sprint start my-launch

The skill handler routes through <bkit-root>/scripts/sprint-handler.js (bkit convention — handlers live at the bkit repo root scripts/ directory, NOT inside skills//scripts/). The handler composes Sprint 3 adapters (state-store + telemetry + doc-scanner + matrix-sync) into Sprint 2 use cases (start / advance / iterate / qa / report / archive). Sprint 1 entities (createSprint / SprintEvents / typedefs) are produced and consumed transparently along the way.

Resolving scripts/sprint-handler.js in this document: throughout this SKILL.md, references to scripts/sprint-handler.js mean <bkit-root>/scripts/sprint-handler.js (the canonical location). LLM dispatchers MUST NOT compose skills/sprint/scripts/sprint-handler.js — that path does not exist (Issue #107, fixed v2.1.19 S2 F2-1).

Arguments

Trust Level Scope (auto-run boundary)

Auto-Pause Safety Pins

Four armed triggers can pause a running sprint:

• QUALITY_GATE_FAIL — M3 > 0 OR S1 < 100
• ITERATION_EXHAUSTED — iter >= 5 AND matchRate < minAcceptable
• BUDGET_EXCEEDED — cumulativeTokens > config.budget
• PHASE_TIMEOUT — phase elapsed > config.phaseTimeoutHours

Pause writes an audit log entry and a SprintPaused event. Resume re-evaluates the triggers and refuses if any are still firing.

Cross-Sprint Architecture (Sprint 1+2+3+4)

USER COMMAND
   v
skills/sprint/SKILL.md (this file — frontmatter triggers in 8 languages)
   v
scripts/sprint-handler.js (English dispatcher)
   v
Sprint 3: lib/infra/sprint -> { stateStore, eventEmitter, docScanner, matrixSync }
   v
Sprint 2: lib/application/sprint-lifecycle -> startSprint / advancePhase / ...
   v
Sprint 1: lib/domain/sprint -> createSprint / SprintEvents / typedefs
   v
DISK: .bkit/state/sprints/<id>.json + .bkit/audit/<date>.jsonl

Examples

See:

• examples/basic-sprint.md
• examples/multi-feature-sprint.md
• examples/archive-and-carry.md

When NOT to Use

• Single-feature PDCA work — use bkit:pdca instead
• Starter level projects — sprint overhead exceeds value
• One-off bug fixes that do not warrant a master plan

Delegation notes

Extended trigger keywords, moved here from the frontmatter description (issue #129 token diet) — one anchor per language stays in the description; the full multilingual list is preserved below:

• JA: スプリント開始, スプリント状態, マスタープラン, マルチスプリント計画, スプリントマスタープラン
• ZH: 冲刺开始, 冲刺状态, 主计划, 多冲刺计划, 冲刺主计划
• ES: iniciar sprint, estado sprint, plan maestro, plan multi-sprint, plan maestro sprint
• FR: demarrer sprint, statut sprint, plan maître, plan multi-sprint, plan maître sprint
• DE: Sprint starten, Sprint Status, Masterplan, Multi-Sprint-Plan, Sprint-Masterplan
• IT: avviare sprint, stato sprint, piano principale, piano multi-sprint, piano principale sprint

Related Skills and Agents

• bkit:pdca — single-feature PDCA cycle (foundation primitive)
• bkit:control — automation level (L0-L4) — surfaces SPRINT_AUTORUN_SCOPE
• bkit:sprint-orchestrator (agent) — full lifecycle coordinator
• bkit:sprint-master-planner (agent) — plan/design generation
• bkit:sprint-qa-flow (agent) — 7-Layer dataFlowIntegrity verifier
• bkit:sprint-report-writer (agent) — KPI + lessons + carry items

10. Skill Invocation Contract (for LLM Dispatchers)

This contract specifies how an LLM dispatcher should construct the args object for each of the 16 sub-actions when invoking the underlying handler via scripts/sprint-handler.js.

10.1 Args Object Schema (per action)

10.1.2 measure action semantics (v2.1.16, Issue #94 F3)

/sprint measure <id> is the user-invokable partial-gate measurement command added in v2.1.16. It routes the requested gate(s) through lib/application/quality-gates/measure-router.js (single SoT shared with sprint-orchestrator self-assessment) and persists results into sprint.qualityGates subject to Trust Level scope.

Three invocation modes (mutually exclusive precedence: gate > gates > phase):

/sprint measure my-launch --gate M4                       # single gate
/sprint measure my-launch --gates M4,M8                   # multi-gate (CSV)
/sprint measure my-launch --phase design                  # phase batch (ACTIVE_GATES_BY_PHASE[design])

Agent routing (Master Plan §11.3 AC4 — 7 gates × 4 agents):

Gates outside this table (M5, M10, S2, S4) return { ok: false, reason: 'unsupported_gate' } — carried to v2.1.17.

Trust Level scope (Master Plan AC5):

• L0 / L1: preview mode — measurement returned but sprint.qualityGates NOT updated, no gate_measured audit entry.
• L2 / L3 / L4: record mode — qualityGates updated + gate_measured audit entry emitted per gate.

Audit emission (when in record mode):

{
  "action": "gate_measured",
  "category": "sprint",
  "actor": "user",
  "target": "<sprintId>",
  "details": {
    "sprintId": "...", "gateKey": "M4", "field": "M4_apiComplianceRate",
    "agent": "gap-detector", "value": 100, "threshold": 95, "passed": true,
    "source": "manual", "phase": "design", "trustLevel": "L3",
    "previousValue": null
  }
}

ENH-292 alignment: multi-gate / phase batch dispatches measurements sequentially (no Promise.all) to avoid #56293 sub-agent caching 10x.

Dispatcher requirement: the LLM dispatcher (main session) must inject deps.agentTaskRunner wrapping Claude Code's Task tool. Without it the use case returns reason: 'no_agent_runner' per gate (deterministic, not silent fail). The handler layer exposes createTaskToolRunner({ invokeTaskTool }) (in scripts/lib/sprint-handler-shared.js, re-exported from scripts/sprint-handler.js) to build this wrapper:

const { createTaskToolRunner } = require('<bkit-root>/scripts/lib/sprint-handler-shared');
const runner = createTaskToolRunner({
  invokeTaskTool: async ({ subagent_type, prompt }) => {
    // delegate to Claude Code's Task tool in the main session
    return { text: await callTaskTool({ subagent_type, prompt }) };
  },
});
await handleSprintAction('measure', { id, gate }, { agentTaskRunner: runner });

Fork mode changes when the result arrives (ENH-478, v2.1.37).

The snippet above assumes callTaskTool resolves to the subagent's finished text. On Claude Code v2.1.232 and later that assumption does not hold in an interactive session: fork mode is on by default, the Agent tool loses its run_in_background parameter, and "a background subagent's results reach Claude as a completion notification in a later turn" (code.claude.com/docs/en/sub-agents). The result is not lost — it arrives on a later turn — but it is not available inside the turn that spawned the subagent.

A dispatcher that awaits it in-turn therefore receives nothing, and the gate reports no_output: an honest "not measured" rather than a wrong score, with the likely cause named in the message. To measure inside one turn, either run non-interactively (-p, where fork mode is off) or set CLAUDE_CODE_FORK_SUBAGENT=0.

Nothing here is a workaround for a defect. It is the shape of the runtime, and a dispatcher that spans turns is the correct adaptation to it.

Two invocation paths:

1. In-process (primary, main session): the LLM dispatcher calls handleSprintAction(...) directly with deps.agentTaskRunner injected. Gate measurement works end-to-end.
2. Subprocess CLI (node scripts/sprint-handler.js ...): runs in a separate Node process that cannot see the Task tool, so it passes {} and gate measurement returns no_agent_runner. Use this path only for non-measurement actions (status, list, help) or when the in-process path is unavailable; for any action that measures gates, use the in-process dispatcher call with an injected runner.

10.1.1 phase --approve semantics (v2.1.16, Issue #95)

When a sprint is at Trust Level L2 (scope.stopAfter = "design") or any other level whose scope.requireApproval blocks a forward transition, the user can re-issue the phase action with --approve (and optional --reason) to cross the scope boundary for this single call only:

/sprint phase my-launch --to do --approve --reason "Design review complete, M4/M8 gates pass"

Semantics (Master Plan §11.2 AC1-AC6):

• Single-use: sprint.autoRun.scope is NOT mutated. The next transition faces the same scope check. To advance through multiple scope-blocking transitions, re-issue --approve each time (or escalate Trust Level via /bkit:control level <N>).
• No trust escalation: sprint.autoRun.trustLevelAtStart and the global automation level (/bkit:control) are unchanged. The approval is recorded per-call.
• Audit-logged: every --approve boundary crossing emits an audit-logger.writeAuditLog({ action: 'scope_boundary_approved', details: { sprintId, from, to, trustLevel, stopAfter, approvedBy, reason } }) entry. The --reason "..." value is the recorded rationale (null when omitted).
• Without --approve the legacy deadlock behavior is preserved: handler returns { ok: false, reason: 'requires_user_approval', stopAfter, hint }.

Use this when you want to advance past the scope boundary for one specific transition (e.g., L2 design → do after design review) without permanently relaxing the trust level.

10.1.1.1 --approve does NOT bypass Quality Gate failures (v2.1.19 S1, CO-S0-6)

Critical semantic clarification (added v2.1.19 S1 in response to S0 discovery of ambiguity — master plan carry-over CO-S0-6):

--approve is the Trust Level scope-boundary escape hatch ONLY. It is NOT a Quality Gate override mechanism.

Why this matters: in v2.1.19 S0 (master plan §23 step 0) we attempted /sprint phase s0-sqm-baseline --to plan --approve and observed { ok: false, reason: 'gate_fail', ... } despite --approve. This is expected behavior — --approve does not satisfy M8 designCompleteness.

Future work (deferred to v2.1.20+): --allowGateOverride flag may be introduced as a gate override (with stronger audit + alarm trail than --approve). Until then, gate failures must be resolved via /sprint measure.

10.1.3 Trust Level Mutation (Persistent) — v2.1.18 (Issue #101)

/sprint trust <sprintId> --to <Level> [--reason "<text>"] [--force]

Mutate the stored sprint.autoRun.trustLevelAtStart for a specific sprint. Unlike --approve (single-use scope boundary override, §10.1.2) or --trustLevel L<N> (per-call volatile override), this command persists the trust level across all subsequent operations on the sprint.

Use cases:

• L1 sprint started conservatively, ready to escalate after design review.
• Demoting L4 sprint to L2 mid-flight after security concern.
• Recovering from L1 "preview-mode lockout" (#101 v2.1.16 root cause — @pruge dandi-village-ledger s1-foundation scenario).

Example:

$ /sprint trust s1-foundation --to L3 --reason "P0 32/32 ready for measurement"
{
  "ok": true,
  "sprintId": "s1-foundation",
  "from": "L1",
  "to": "L3",
  "reason": "P0 32/32 ready for measurement",
  "actor": "user",
  "forced": false,
  "trustScoreAtMutation": null,
  "blastRadius": "low",
  "auditEntryId": "..."
}

$ /sprint measure s1-foundation --gate M1
{ "trustLevel": "L3", "mode": "record", "value": 92.3, ... }  # ✦ now record mode

Downgrade Guardrail:

Major downgrades (≥2 levels, e.g. L4 → L2 or L3 → L1) require:

• trustScore >= 80 (from .bkit/state/trust-profile.json trustScore field — 6-component weighted sum: pdcaCompletionRate 0.25 / gatePassRate 0.2 / rollbackFrequency 0.15 / destructiveBlockRate 0.15 / iterationEfficiency 0.15 / userOverrideRate 0.1), OR
• --force flag (explicit override + forced: true audit + blastRadius: 'high' for Defense Layer 6 alarm).

Minor downgrades (1-level diff, e.g. L3 → L2) are not blocked.

Idempotent Path:

from === to (e.g. --to L3 when sprint already at L3) returns { ok: true, noop: true } and also emits audit with noop: true field (CTO §C3 review: monitoring blind-spot prevention — surfaces automation patterns hitting idempotent paths).

Actor Auto-Detection (CTO §E6 spoofing mitigation):

actor field is auto-detected:

• explicit args.actor (if 'user'|'agent'|'system'), else
• process.env.CLAUDE_AGENT_ID set → 'agent', else
• default 'user'.

Audit:

Every mutation (including no-op) emits an audit-logger entry:

{
  "action": "sprint_trust_changed",
  "category": "sprint",
  "actor": "user",
  "target": "s1-foundation",
  "targetType": "feature",
  "blastRadius": "low",
  "details": {
    "sprintId": "s1-foundation",
    "from": "L1",
    "to": "L3",
    "reason": "...",
    "trustScoreAtMutation": null,
    "forced": false,
    "noop": false,
    "actor": "user",
    "timestamp": "2026-05-21T..."
  }
}

Comparison Table:

10.2 Trust Level Acceptance

All actions that accept a Trust Level recognize three input forms (handled by normalizeTrustLevel in scripts/sprint-handler.js):

• args.trustLevel (preferred, explicit handler arg)
• args.trust (CLI --trust L3 natural mapping)
• args.trustLevelAtStart (stored property leak; defensive only)

Precedence: trustLevel > trust > trustLevelAtStart. Defaults to L2 when none provided or value is invalid (case-insensitive match against L0-L4).

v2.1.19 S1 F1-4 default change: default lowered from L3 to L2 per Safe Defaults principle (master plan §3.2 Controllable AI Principles). The handler now aligns with lib/domain/sprint/entity.js createSprint which already defaulted to L2 — eliminates the v2.1.16~v2.1.18 drift between handler default (L3) and entity default (L2).

--trust L1 explicit warning: when the user explicitly requests L1 at /sprint init, the handler emits a stderr warning + audit sprint_trust_warning event re: preview-mode lockout risk (v2.1.18 #101 follow-up). The warning is education-only — L1 sprint init still succeeds.

10.3 Natural Language Mapping Rules

When the user invokes the skill with mixed slash command + natural language (e.g., /sprint start S1-UX Phase 1 PRD please proceed thoroughly), the LLM dispatcher SHOULD:

1. Extract action: first non-flag token after /sprint → action.
2. Extract id (kebab-case): scan remaining tokens for the first kebab-case identifier (matches /^[a-z][a-z0-9-]{1,62}[a-z0-9]$/). Lowercase if needed. Example: S1-UX → s1-ux.
3. Disambiguate via AskUserQuestion: if multiple kebab-case candidates or none, prompt the user to confirm the intended sprint id.
4. Load name from state: for start action on an existing sprint, the name field can be resolved by handleStatus({ id }) first; otherwise fall back to the id itself.

10.4 Example — Resume Existing Sprint

User: /sprint start s1-ux
LLM dispatch:
  1. action = "start", id = "s1-ux"
  2. status = await handleSprintAction("status", { id: "s1-ux" })
  3. name = status.sprint.name  // "S1-UX P0/P1 Quick Fixes"
  4. await handleSprintAction("start", { id: "s1-ux", name })
  5. Handler invokes load-then-resume path (P0 fix) — phase preserved

10.5 Example — Ambiguous Natural Language

User: /sprint start S1-UX Phase 1 PRD proceed thoroughly
LLM dispatch:
  1. action = "start"
  2. Candidates: ["s1-ux"]  (kebab-case extracted from "S1-UX")
  3. AskUserQuestion: "Did you mean to start sprint 's1-ux' and continue
     with Phase 1 (PRD)?" → user confirms
  4. await handleSprintAction("start", { id: "s1-ux", ... })

10.6 Error Handling

Handler returns { ok: false, error: <string>, ... } on failure. LLM dispatcher SHOULD surface the error verbatim to the user and offer remediation (e.g., for error: 'Sprint not found', suggest /sprint list).

10.7 CLI Mode (P1 fix)

The same handler is invokable as a standalone CLI when run as node scripts/sprint-handler.js <action> [id] [--flags]. Useful for headless tests, debugging, and CI integration. The CLI parser accepts --key value and --key=value forms, with the first positional argument after action treated as id if no --id flag is provided.

Exit codes: 0 (success), 1 (handler returned ok: false), 2 (exception thrown).

11. Master Plan Generator (16th Sub-Action)

The master-plan action generates a multi-sprint roadmap via the bkit:sprint-master-planner agent (isolated subagent spawn) and persists both markdown documentation and state JSON.

11.1 Workflow

USER: /sprint master-plan q2-launch --name "Q2 Launch" --features auth,payment
   |
SKILL.md dispatches -> scripts/sprint-handler.js handleMasterPlan
   |
handleMasterPlan calls lib/application/sprint-lifecycle/master-plan.usecase.js generateMasterPlan
   |
generateMasterPlan validates input + loads existing state (idempotent check)
   |
If deps.agentSpawner provided: spawn bkit:sprint-master-planner agent -> markdown
If not: dry-run via templates/sprint/master-plan.template.md substitution
   |
Atomic write: .bkit/state/master-plans/<projectId>.json (state first)
   |
File write: docs/01-plan/features/<projectId>.master-plan.md (markdown)
   |
Audit: lib/audit/audit-logger.js writeAuditLog({ action: 'master_plan_created' })
   |
Optional Task wiring: deps.taskCreator called N times for N sprint tasks

11.2 Idempotency + Force Overwrite

• Default: idempotent. Second call with same projectId returns existing plan.
• --force flag: overwrites both state JSON and markdown. Audit entry has details.forceOverwrite: true.
• Audit ACTION_TYPE remains 'master_plan_created' for both cases (PM-S2G).

11.3 Dry-Run vs Agent-Backed Generation

When the caller (LLM dispatcher at main session) does NOT inject deps.agentSpawner, the use case generates a minimal valid markdown by substituting variables in templates/sprint/master-plan.template.md. The output is a skeleton — header, context anchor placeholders, empty features table, empty sprints array. This dry-run mode is useful for unit tests and when the user wants a starting template to fill manually.

When deps.agentSpawner is injected, the use case calls it with { subagent_type: 'bkit:sprint-master-planner', prompt: <built> } and uses the returned output field as the markdown content.

11.4 State Schema v1.0

The state JSON at .bkit/state/master-plans/<projectId>.json:

{
  "schemaVersion": "1.0",
  "projectId": "q2-launch",
  "projectName": "Q2 Launch",
  "features": ["auth", "payment", "reports"],
  "sprints": [],
  "dependencyGraph": {},
  "trustLevel": "L3",
  "context": { "WHY": "", "WHO": "", "RISK": "", "SUCCESS": "", "SCOPE": "" },
  "generatedAt": "2026-05-12T20:00:00Z",
  "updatedAt": "2026-05-12T20:00:00Z",
  "masterPlanPath": "docs/01-plan/features/q2-launch.master-plan.md"
}

The sprints array is populated by the S3-UX context-sizer.js use case. S2-UX leaves it as an empty stub.

11.5 Task Management Integration (Optional)

When the caller injects deps.taskCreator, the use case iterates plan.sprints sequentially (ENH-292 caching alignment) and calls deps.taskCreator(...) once per planned sprint with addBlockedBy populated from the previous sprint's task ID. This enables automatic Task list creation for multi-sprint roadmaps.

When deps.taskCreator is undefined or plan.sprints.length === 0, Task creation is silently skipped (no error).