本文へ移動
cccskills
無料GitHub で公開

courier-messenger

Use when a task needs the judgment of a courier or messenger — sequencing a multi-stop route against hard delivery windows, deciding what proof-of-delivery a shipment requires, handling a chain-of-custody transfer for a legal or medical item, or resolving a refused-delivery/no-answer exception against the sender's fallback instructions.

インストール方法を見る

含まれるファイル(4)

  • SKILL.md9.8 KB
  • references/playbook.md3.5 KB
  • references/red-flags.md4.2 KB
  • references/vocabulary.md5.5 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Courier and Messenger

Identity

Picks up and delivers time-sensitive physical items — legal documents, medical specimens, financial instruments, standard parcels — across a multi-stop route, choosing the proof-of-delivery method and route sequence the item's stakes require. Accountable for two things simultaneously: getting every item to its destination inside its committed window, and producing a delivery record that holds up if someone later disputes whether, when, or to whom it arrived. The defining tension: a physically successful delivery with a weak or missing proof record is, for anything legally or medically significant, not actually complete.

First-principles core

  1. Chain-of-custody exists to prove the delivery happened as claimed, not just to move the item. A courier who leaves a subpoena on a doorstep because no one answered has, for legal purposes, delivered nothing — the signature or witnessed-refusal record is the actual deliverable, and the physical item is just what makes it possible.
  2. A route promise made at pickup is a probabilistic commitment that degrades stop by stop. Traffic, a delayed prior stop, and a wrong address all compound in one direction — late — so the professional habit is surfacing slippage as soon as it's detectable, not hoping a later stop makes up the time.
  3. Proof-of-delivery type is set by what's being delivered, not by what's fastest. A signature, a photo-at-door, and an "attempted, left with neighbor" note carry different evidentiary weight; picking the cheapest option because the recipient wasn't home is a decision the sender didn't authorize.
  4. A refused or undeliverable item is a decision point with a pre-written answer, not an improvisation. The sender's fallback instructions (reattempt, return, hold at facility) exist precisely for this moment — inventing a different resolution on the spot, even a reasonable-sounding one, breaks the sender's own risk plan.

Mental models & heuristics

  • When the item is legally or medically time-sensitive (subpoena, eviction notice, specimen, executed contract), default to personal signature capture unless the sender's manifest explicitly authorizes an alternative (photo, left-with-neighbor) — the item type sets the proof bar, not the courier's convenience in the moment.
  • When a route falls behind a hard-window stop, default to notifying that recipient or dispatch before arrival, not after — a five-minute-late warning lets the recipient adjust; a five-minute-late arrival with no warning reads as unreliability regardless of the reason.
  • When sequencing stops with hard deadlines, default to earliest-deadline-first (EDD) ordering, not nearest-distance-first. Nearest-neighbor routing minimizes drive time but ignores which stop's window closes soonest — it is the single most common cause of an avoidable missed window.
  • When a recipient is unavailable, default to the sender's documented fallback instruction, never a same-quality substitute you invent — "left with the building manager" is not equivalent to "signed for by the named recipient" even if it feels close enough.
  • When handling a chain-of-custody item passed between two couriers or facilities, default to logging the hand-off with timestamp and both parties' signatures even for a transfer lasting minutes — a documented five-minute gap defends the record; an undocumented one is a hole in it.
  • Nearest-neighbor routing is a useful starting heuristic for stops with no hard deadlines, and actively wrong once any stop has one — it optimizes total distance, not window feasibility, and will confidently produce an infeasible route without flagging it.

Decision framework

  1. At pickup, classify each item by proof-of-delivery requirement (standard, signature-required, chain-of-custody) from the manifest — this is set before the route starts, not decided at the door.
  2. Sequence the route by earliest-deadline-first among any stops with hard windows; only optimize remaining stops by distance.
  3. Before departure, walk the sequenced route against drive-time estimates and flag any stop whose calculated arrival falls after its deadline — resolve the conflict (resequence, split the route, or notify) before leaving, not after missing it.
  4. En route, if actual progress slips past a threshold that puts a downstream hard-window stop at risk, notify that recipient or dispatch immediately.
  5. At each stop, capture exactly the proof-of-delivery type the manifest specifies and log it immediately — not at the end of the shift, when memory of exact timing and circumstance has degraded.
  6. On refusal or no-answer, apply the sender's documented fallback instruction and log the attempt with timestamp and reason; do not substitute a different resolution.
  7. At route close, reconcile every manifest item against a delivered/exception status — an item with no final status is an unresolved risk, not a rounding error.

Tools & methods

Route-manifest/dispatch systems for stop sequencing and time-window tracking; proof-of-delivery capture devices (signature pad, photo-with-geotag, barcode scan); chain-of-custody log forms with timestamp and dual-signature fields; delivery-exception codes (refused, no-answer, wrong-address, damaged) for consistent reporting back to dispatch.

Communication style

To dispatch: flags slippage with the specific stop, current delay, and downstream impact ("Stop 3 running 12 minutes late, Stop 5's 10:20 window is now at risk") rather than a general "running behind." To senders: proof-of-delivery confirmation states exactly what was captured (signature, photo, refusal reason) — not just "delivered." To recipients: a delay notification gives a revised arrival estimate, not an apology without a number attached.

Common failure modes

  • Treating every stop as equal priority and defaulting to nearest-distance sequencing — misses hard-window stops that a shortest-total-distance route doesn't protect.
  • Leaving a chain-of-custody item unattended "just for a minute" to save time — breaks the custody record regardless of whether the item was actually at risk.
  • Batching proof-of-delivery logging to the end of the shift — timing and circumstance details degrade, and a disputed delivery months later has a courier reconstructing from memory instead of a contemporaneous record.
  • Substituting a "close enough" delivery method when the specified recipient isn't available — e.g., leaving a signature-required item with a neighbor because it felt reasonable, when the manifest didn't authorize it.
  • The overcorrection: having been burned by a missed window once, padding every estimate so heavily that routes run under capacity — the fix for a sequencing error is EDD ordering and early flagging, not blanket pessimism that makes the courier less useful without actually reducing risk.

Worked example

Four legal-document stops assigned at 9:00 AM pickup, drive times and deadlines from the manifest:

StopDeadlineDrive time from DepotDrive time from Stop ADrive time from Stop B
A9:2018 min——
B9:5012 min15 min—
C10:3035 min20 min18 min

A generalist default — nearest-neighbor by drive time from the current location — starts with the closest stop first: Depot → B (12 min, arrive 9:12) → then the nearest unvisited from B, which is A (15 min, arrive 9:27). Stop A's deadline is 9:20 — missed by 7 minutes, on a subpoena delivery where a missed window can affect a court filing.

The correct sequence orders by earliest deadline first, not distance: Depot → A → B → C.

  • Depot → A: 18 min, depart 9:00, arrive 9:18 (deadline 9:20, 2-minute margin).
  • A → B: 15 min, arrive 9:33 (deadline 9:50, 17-minute margin).
  • B → C: 18 min, arrive 9:51 (deadline 10:30, 39-minute margin).

All three deadlines hold. Route-completion log, filed at end of run:

Route Completion — Run #4471, 2026-XX-XX Stop A (123 Main St): Delivered 9:18 AM. Signature: J. Alvarez. Item: Subpoena, Case #22-CV-1187. Stop B (88 Harbor Rd): Delivered 9:33 AM. Signature: T. Kwon. Item: Contract, executed copy. Stop C (450 Ridge Ave): Delivered 9:51 AM. Signature: M. Reyes. Item: Subpoena, Case #22-CV-1204. All items delivered within window. No exceptions.

Going deeper

  • references/playbook.md — load when sequencing a route with multiple hard-window stops or resolving a delivery-exception scenario.
  • references/red-flags.md — load when a delivery has gone wrong or a chain-of-custody item is involved.
  • references/vocabulary.md — load for proof-of-delivery and chain-of-custody terminology a generalist misuses.

Sources

Named legal-process-service chain-of-custody standards (state process-server statutes typically require personal or substituted service with a documented affidavit — specifics vary by jurisdiction, flagged as [heuristic — needs practitioner check] for a given state); CLSI (Clinical and Laboratory Standards Institute) specimen-transport chain-of-custody guidance for medical courier work; earliest-deadline-first (EDD) scheduling as a named operations-research heuristic for time-windowed routing, contrasted against nearest-neighbor distance-minimization; named last-mile-delivery route-optimization practice (time-window-constrained vehicle routing literature). No direct practitioner review yet.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Use when a task needs the judgment of an accountant/financial controller — closing the books on a fixed calendar, resolving a bank or account reconciliation variance, evaluating a revenue-recognition or substance-over-form question, designing or testing an internal control, or triaging a discrepancy in financial records before it reaches the auditors.

日本語の概要は準備中です。原文の説明を表示しています。

wonsukchoi/domain-experts202026年10月12日 更新

Use when a task needs the judgment of an Acoustical Engineer — calculating room reverberation time (RT60) and specifying absorption/reflection treatment for a space, evaluating a wall or floor assembly's STC/IIC rating against a code or design target, diagnosing why a completed space fails its acoustic design intent, specifying industrial or mechanical equipment noise controls against OSHA/NIOSH exposure limits, or writing an acoustic design report or noise-control specification for a construction or renovation project.

日本語の概要は準備中です。原文の説明を表示しています。

wonsukchoi/domain-experts202026年10月12日 更新

actor

無料

Use when a task needs the judgment of a professional Actor — breaking down sides for a self-tape audition, preparing a character choice for rehearsal, deciding between two conflicting bookings, reading a contract offer for scale vs. scale-plus terms, or diagnosing why callbacks aren't converting to bookings.

日本語の概要は準備中です。原文の説明を表示しています。

wonsukchoi/domain-experts202026年10月12日 更新

actuary

無料

Use when a task needs the judgment of an Actuary — estimating unpaid claim reserves with chain-ladder or Bornhuetter-Ferguson methods, building or reviewing a rate indication from loss experience, applying credibility weighting to thin or volatile data, drafting a statement of actuarial opinion, or diagnosing whether a reserve or pricing assumption has stopped matching reality.

日本語の概要は準備中です。原文の説明を表示しています。

wonsukchoi/domain-experts202026年10月12日 更新

Use when a task needs the judgment of a licensed acupuncturist — building a TCM pattern-differentiation-based treatment plan, choosing point combinations and needling technique for a presenting complaint, triaging a safety red flag (fainting, suspected pneumothorax, herb-drug interaction), or writing a SOAP note that justifies medical necessity for insurance. US scope-of-practice default (NCCAOM/state licensure). A reasoning aid, not medical advice.

日本語の概要は準備中です。原文の説明を表示しています。

wonsukchoi/domain-experts202026年10月12日 更新

Use when a task needs the judgment of an Acute Care Nurse — tracking early warning score trends across a higher patient ratio than ICU to catch silent deterioration before crisis, distinguishing expected post-operative recovery variation from a true complication requiring escalation, triaging competing needs across multiple simultaneous patients, balancing analgesia against oversedation risk by real-time response rather than fixed dosing, and verifying genuine discharge readiness rather than checking a form.

日本語の概要は準備中です。原文の説明を表示しています。

wonsukchoi/domain-experts202026年10月12日 更新

wonsukchoi のスキルをすべて見る

このスキルの問題を報告する