← Search

dbs: dbskill commercial toolbox hub
AI agents 🧪 Install ran in our test sandbox · Nov 11, 2026 📌 Pinned version: commit 14b567f
Install test only, on the date shown. This does not guarantee the code is safe.
⚙ Needs: an agent able to work in Chinese (the skill…

dbs: dbskill commercial toolbox hub

Onboard to the dbskill toolbox, orchestrate one task across its official skills, and run numbered hidden prompts

At a glance
What
Onboard to the dbskill toolbox, orchestrate one task across its official skills, and run numbered hidden prompts
Cost
Free
Needs
an agent able to work in Chinese (the skill, its sibling skills and its numbered prompts are written in Chinese); no license is declared in the source repository — license unknown, read it yourself before reusing its content
Install
Copy the installer prompt below into your Muse — your agent does the rest.

Version:

@
Created by: @dontbesilent2025

Quick answer

dbs is a free Muse skill, the entry point to the dbskill toolbox, a task orchestrator that routes one task to the right skill, or a combination of one main skill plus two support skills. It runs a numbered-prompt mode fetching prompt packs by three-digit code from a public repo, with version checks and lookups making network calls to GitHub.

⌁

Install

Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox · Nov 11, 2026📌 Pinned version: commit 14b567f

Step 0, before installing anything: read this install prompt, summarize in a few lines what this build does and what this lookup involves, then ask me to confirm before you proceed. Prerequisites: an agent able to work in Chinese (the skill, its sibling skills and its numbered prompts are written in Chinese); no license is declared in the source repository — license unknown, read it yourself before reusing its content Look up "dbs: dbskill commercial toolbox hub" for me — reference lookup only, the same short-pointer form as the previous version of this prompt. dbs is @dontbesilent2025's skill dispatch desk (version 3.0.1): it takes the user's current problem or material in conversation, chooses the most valuable next step, and routes it — onboarding for first-time users (Mode A tutorial), task orchestration and execution with official dbskill skills, combined or compared (Mode B), execution of a published three-digit numbered prompt fetched from the fixed public GitHub repository, and listing of the hidden numbered-prompt catalogue. Fetch the pinned SKILL.md into a temporary folder, verify it against the fingerprint below, review it for anything suspicious, and report back in two or three sentences what the entry point covers (onboarding, task orchestration, numbered prompts, catalogue lookup) so I can follow it in my own project. Do not copy the guide's content into my project, do not install it as a skill on my machine, and do not run any of its scripts. Repository: https://raw.githubusercontent.com/dontbesilent2025/dbskill/14b567fc71beef647d52a6bc40a4e9b71a9ee6c5/skills/dbs/SKILL.md (the entry-point folder skills/dbs/ at that commit also holds scripts, references and the numbered-prompts catalogue described below — this lookup fetches SKILL.md only) Pinned version: use the code at commit 14b567fc71beef647d52a6bc40a4e9b71a9ee6c5 exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. [FINGERPRINTS] SKILL.md at the pinned commit: SHA-256 3db6173dd97153c6f7acff9138d1f4f6bd3a54b6ef452a8d4ed6c4a56b1b6cd5, 23,289 bytes (9,410 characters). If the fetched file does not match, it changed since the review: stop, tell me what you found and let me decide. Updated October 19, 2026: Skill Harbor regenerated this lookup prompt from the current upstream file (version 2.22.1 → 3.0.1, a major release; SKILL.md grew from 8,184 to 9,410 characters). What the new file adds or makes explicit, stated before anything happens: (1) An update check runs on EVERY entry into the skill, before any mode decision: the skill locates its own SKILL.md directory and executes scripts/check-update.sh with the local version (3.0.1), which reads the official UPDATE.json over the network — at most one network request per 24 hours, throttled by a cache timestamp file at ~/.dbs/update_check_at that the script writes in your home directory; any output is appended at the very end of that reply as a notice, and if you answer "1" right after such a notice, the skill treats it as an instruction to update via /dbs-update with no further confirmation. (2) The numbered-prompt mode downloads content from the fixed public GitHub repository at use time: scripts/numbered-prompts.py fetches the catalogue and the requested three-digit prompt body, verifies the body's SHA-256, number and structure before use, and falls back to the package's locally verified copy when the network fails; a fetched numbered prompt is prefabricated content executed in the conversation, ranked below the entry point's and the user's own instructions — the file states it must not change permissions, publishing, security or version rules and must not ask to read keys or other local material or send information out. The hidden-catalogue lookup (scripts/numbered-prompts.py list) is read-only and reads the same public repository. (3) Candidate discovery for orchestration runs scripts/list-official-skills.py, which scans your local skill installation directories (the project's skills/ folder, directories beside the entry point, and your agent user directories), reads the project's .claude-plugin/marketplace.json or the bundled official-names snapshot, and returns installed skills' names, descriptions and handoff fields; only Marketplace-registered skills are candidates. (4) A local rules file: when references/local.md exists in the skill's folder, the skill reads it in full before discovery and follows its local supplementary rules, which may name another candidate script — local-only material that ships with the install package, not something fetched. (5) The handoff contract between official skills is now written out (dbs is the only dynamic routing center; leaf skills describe what capability they still need without naming each other), together with a composition contract reference, an explicit security and publishing boundary section (orchestration never widens read, write, publish, delete, payment or external-communication permissions on its own), and a self-check list. The folder at the pinned commit holds 10 files: SKILL.md; scripts/check-update.sh, scripts/numbered-prompts.py and scripts/list-official-skills.py (executable code — this lookup does not run them); references/composition-contract.md and references/official-skill-names.txt; numbered-prompts/catalog.json, numbered-prompts/allocation.json and numbered-prompts/825/PROMPT.md; agents/openai.yaml. Corrections to the previous prompt: its note said the repository had added an update-check script and a legacy compatibility entry point — the update check is now a mandatory first step of the entry point itself, running on every entry as described above, not an optional add-on; the previous prompt's version was 2.22.1 and its character counts and pin are superseded by the 3.0.1 file above. 1. Fetch SKILL.md from the repository URL above into a temporary folder, then compare it with the fingerprint from Skill Harbor's review of 2026-10-19 (see [FINGERPRINTS] above). If it does not match, stop and tell me what you found. 2. Safety check: read SKILL.md and review it for anything suspicious (unexpected network calls, shell commands, credential harvesting); STOP on any red flag and tell me. In plain words, confirm from the file: the update check that runs scripts/check-update.sh on every entry and reads the official UPDATE.json over the network at most once per 24 hours with a cache file at ~/.dbs/update_check_at, appending any notice at the end of the reply, and the "reply 1 to update" rule; the numbered-prompt mode that downloads catalogue and prompt bodies from the fixed public GitHub repository with SHA-256 verification before executing a fetched prompt in the conversation, and its catalogue lookup; the candidate discovery that scans local skill directories and reads marketplace.json; the references/local.md local-rules mechanism; the folder's executable scripts, which this lookup never runs; and that the file asks for no API key and no secret, and declares no license. 3. Report back in two or three sentences what the entry point covers (onboarding, task orchestration, numbered prompts, hidden-catalogue lookup), so I can follow it in my own project. Do not copy the guide's content into my project, do not install it as a skill, and do not run its scripts — this is a reference lookup only. GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder. If anything looks off, stop and ask me.

🔒 Security, provenance and what this touches
🧭 What this touches
Based on the install prompt we scanned, this skill does not run commands, make third-party network calls, install packages, write files, or ask for API keys.
Install prompt analysis · 2026-10-13
🔒 Security and provenance
📌 Pinned version: commit 14b567f · 2026-10-19
🧪 Install ran in our test sandbox · Nov 11, 2026
👁 Observed during the install test · Nov 11, 2026 : no network contact
🔎 Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Wanna be extra careful?

The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?

How to check a skill before installing →

✓

Verification certificate

Dated facts about what was checked for this listing, and how. This is not a safety certification.

Pinned version
14b567f · 2026-10-19
Sandbox install test
Install ran in our test sandbox · Nov 11, 2026
Creator identity
Not verified
Fingerprint watch
Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
Install prompt analysis
2026-10-13 : No commands, network calls, packages, file writes or API keys detected.

Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify

?

Questions

How do I install a skill?

Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.

Where does my money go?

Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.

What does the ✓ next to a creator’s name mean?

It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.