Index of every page on this site: llms.txt
All skills
open-neko avatar

/support-desk

@621f131
by open-nekoopen-neko/openneko22 stars
3
Run it yourself

Model or interpret a generated customer-support app, especially for Zendesk-shaped data. Use when a user wants a support desk, ticket and requester tracking, Zendesk replacement or import, queues and assignment, service priorities or statuses, or help interpreting a support import report.

  • 2 files
  • 3.8 KB
  • Updated 3 months ago
  • GitHub
Use this Skill: https://skilld.dev/gh/open-neko/openneko/support-desk

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈76 tokens for metadata: the name and description. ≈839 when used: this file. ≈51 more on demand in 1 file.

Description uses 3.8% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈304 characters, or 3.8%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Support Desk

Apply support-domain knowledge through the generic records engine. Use mcp_neko_records_browse_blueprints with blueprint: "support" to load the canonical starting payload. Before proposing or changing the app, read ../app-builder/SKILL.md completely and follow it to adapt the loaded payload. Before reading or writing support rows, read ../records/SKILL.md completely. The blueprint is a prior, not a fixed ticketing product.

Model the service workflow

  • Organization is the customer company or service account.
  • Requester is the person asking for help and may reference an organization.
  • Ticket is the unit of triage, ownership, status, priority, and SLA timing.
  • Comment is the durable conversation entry. Keep author, public/internal visibility, body, and creation time when the source provides them.
  • Add queues, products, incidents, or custom objects only when the operator's routing and reporting workflow needs them.

Use reference fields for requester, organization, ticket, and queue links; textarea for descriptions/comment bodies; datetime for opened, due, and resolved timestamps; and picklists for status, priority, type, and channel.

Status, priority, and assignment

The canonical status prior is new → open → pending → solved → closed, and the priority prior is low / normal / high / urgent. Adapt these values to the operator's actual lifecycle. Stable values are permanent data; labels may be friendlier. Preserve imported unknown values for explicit mapping rather than silently treating them as open or normal.

Ticket assignee or owner language normally maps to owner_user_id with visibility: owner when members should see only their assigned tickets. Queues are business references, not a replacement for actor ownership. Do not guess an OpenNeko user for an unlinked source assignee.

Keep internal comments distinguishable from requester-visible replies. Do not promise outbound email, SLA escalation, or automations merely because their fields exist; those are separate governed workflows.

Zendesk-shaped imports

Map common source concepts as follows, then adapt:

  • organization → organization
  • user with requester role → requester
  • ticket → ticket
  • ticket comment or audit body → comment
  • group → queue when queue routing matters

Keep connector instance plus source ID so identical IDs from separate Zendesk accounts never collide. Preserve requester and organization rows even when a legacy ticket has a dangling relationship; report the integrity issue instead of inventing a target.

Baseline CSV import is insert-only. In an import report, distinguish inserted, duplicate, rejected, unresolved reference, and unlinked/conflicting owner counts. Never describe duplicates as updates or assume a rejected comment means its parent ticket failed too.

Do not claim a CSV load is a live mirror. Connected mirror mode stays locally read-only until an explicit final delta, watermark verification, and cutover.

Working with support records

Use $records for every read or write. Resolve the exact ticket/requester/ organization/comment and any reference IDs before proposing changes. For ambiguous tickets, disambiguate with ticket ID, requester, subject, status, creation time, or organization; never choose the first similar subject.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at 621f131. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 4 weeks ago.

Activeupdated 3 months ago
type
Skill

README badge

README badge for open-neko/openneko/support-desk