Agent skill

lp-weekly

S10 weekly orchestration wrapper. Coordinates the full weekly decision, audit/CI, measurement compilation, and experiment lane flows into one deterministic sequence. Additive-first; never replaces canonical S10 authorities. No new hard stage blocks in first iteration.

View SKILL.md on GitHub Repository
Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/lp-weekly

SKILL.md

lp-weekly — S10 Weekly Orchestrator

/lp-weekly is the S10 orchestration entrypoint. It coordinates existing weekly flows into one deterministic sequence and emits an additive weekly packet. It does NOT replace stage graph authority or weekly decision-prompt authority.

Authority Stack

Surface Authority Non-authority
docs/business-os/startup-loop/specifications/loop-spec.yaml Stage graph, stage ordering Not weekly memo content
docs/business-os/workflow-prompts/_templates/weekly-kpcs-decision-prompt.md Weekly decision content (Sections A–H) Not stage graph
/lp-weekly Execution order, preflight checks, lane sequencing, packetization, cross-artifact linking Does NOT replace prompt authority or stage graph authority

Invocation

/lp-weekly --biz <BIZ> --week <YYYY-Www> [--as-of-date <YYYY-MM-DD>] [--run-root <path>]

Parameters

Parameter Required Default Notes
--biz Required — Business identifier (e.g., BRIK, PET). Must match a directory under docs/business-os/strategy/.
--week Required — ISO week key, UTC Monday anchor. Format: YYYY-Www (e.g., 2026-W07). Derived from invocation date if not supplied; operator may override.
--as-of-date Optional Today (YYYY-MM-DD) Used for artifact naming and week key derivation.
--run-root Optional docs/business-os/strategy/<BIZ>/ Path prefix for stage artifacts. Deterministically derived from --biz when omitted.

Week key derivation rule

Derived from UTC date at invocation using ISO 8601 week numbering. Week boundary: Monday 00:00:00 UTC to Sunday 23:59:59 UTC. Zero-padded two-digit week number, e.g., 2026-W07.

Global Invariants

Operating mode

ORCHESTRATE + SEQUENCE + EMIT

Allowed actions

  • Read stage artifacts, KPI sources, prior-week packet, and canonical prompt docs.
  • Invoke sub-flows: GATE-LOOP-GAP-03 feedback-loop audit, /lp-experiment readout (per experiment), KPCS prompt handoff.
  • Write one weekly packet per YYYY-Www key (overwrite in-place on rerun).
  • Emit lane status summaries and REM delta sections within the packet.

Prohibited actions

  • Replacing or modifying the canonical weekly KPCS decision memo.
  • Introducing new hard stage blocks or gate-semantics changes (first iteration no-block posture).
  • Modifying docs/business-os/startup-loop/specifications/loop-spec.yaml stage graph.
  • Removing or suppressing GATE-BD-08 or GATE-LOOP-GAP-03 dispatch behaviors.
  • Silent failure: partial inputs must produce restricted lane state + REM, not silent degraded output.
  • Destructive shell or git commands.
  • Creating files outside the weekly packet and its cross-artifact links.

Module Routing

Execute modules in sequence:

Step Module Sequence role
1 modules/preflight.md Validate inputs; determine ready or restricted; emit missing-input list
2–5 modules/orchestrate.md Compile measurements; execute decisioning handoff; run audit/CI lane (a); roll up experiments lane (c)
6 modules/publish.md Emit additive weekly packet; update latest pointer; record lane statuses

See each module for detailed sub-step sequencing.

Lane Summary

Lane ID Sequence steps Exit
Audit + CI a Steps 3–4 (orchestrate.md) complete or incomplete
Measurement + reporting b Step 2 (orchestrate.md) ready or restricted
Next experiments c Step 5 (orchestrate.md) complete or carry-forward

Idempotency

Packet path is fixed per week key: docs/business-os/strategy/<BIZ>/s10-weekly-packet-<YYYY-Www>.md. On rerun for the same week key: overwrite in-place. No version-suffix copies. Git history is the audit trail. All canonical artifact references are re-linked in the overwritten packet; they are never deleted.

Output Contract

Weekly packet: docs/business-os/strategy/<BIZ>/s10-weekly-packet-<YYYY-Www>.md

Required packet sections:

  • week_key: YYYY-Www
  • biz: business identifier
  • as_of_date: invocation date
  • preflight_status: ready or restricted
  • missing_inputs: list (empty if ready)
  • lane_a_status: complete | incomplete
  • lane_b_status: ready | restricted
  • lane_c_status: complete | carry-forward
  • kpcs_decision_ref: path to canonical KPCS decision artifact
  • feedback_audit_ref: path to feedback loop audit artifact (if exists this cycle)
  • measurement_summary: normalized measurement summary or restricted note
  • rem_delta: new and resolved REM items this week
  • experiment_portfolio_summary: rollup of active/completed experiments
  • next_cycle_backlog_delta: new next-experiment specs added or carried forward
  • pending_reviews_delta: new R9 flags, review items, stale queue packets, and resolved items found by the Lane A pending reviews scan (Step 4.8)
  • run_trace: list of sub-flows invoked with their outcomes

Restricted lane state: when lane b inputs are incomplete, lane_b_status is set to restricted. A REM task is emitted identifying the missing denominator or KPI family. The packet is still written; the operator is not blocked from closing the weekly cycle.

Exit Conditions

Condition Description Packet emitted?
complete All lanes reached exit or documented carry-forward; packet written Yes
partial Preflight passed but one or more lanes are incomplete or restricted; REM emitted per incomplete lane Yes
restricted Preflight returned restricted; critical inputs missing; packet written with preflight_status: restricted Yes (restricted)
fail-closed --biz does not exist, or --week cannot be derived, or run_root is missing No

Integration

Sub-flows invoked by /lp-weekly:

  • GATE-LOOP-GAP-03 feedback-loop audit — pre-decision advisory input (lane a)
  • KPCS decision prompt handoff — authoritative weekly decision (step 3, decisioning)
  • /lp-experiment readout — per-experiment readout (lane c, experiment rollup step)

Startup-loop integration (Phase 1): During S10 advance, /startup-loop routes through /lp-weekly before returning to standard BOS sync and advance flow. See .claude/skills/startup-loop/modules/cmd-advance.md for Phase 1 dispatch invocation.

Canonical artifacts produced outside /lp-weekly (referenced, not replaced):

  • docs/business-os/strategy/<BIZ>/<YYYY-MM-DD>-weekly-kpcs-decision.user.md
  • docs/business-os/strategy/<BIZ>/feedback-loop-audit-<YYYY-MM-DD>.md
  • docs/business-os/strategy/<BIZ>/<YYYY-MM>-monthly-audit.user.md

Orchestration Contract

Full S10 sequence, lane contracts, authority isolation, and idempotency rules: docs/business-os/startup-loop/contracts/s10-weekly-orchestration-contract-v1.md

Recommended Agent Skills

Expand your agent's capabilities with these related and highly-rated skills.

majiayu000/claude-skill-registry

agent-ops-spec

Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-state

Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-spec

Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-testing

Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-testing

Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-state

Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.

163 31
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results