Skip to content
SkillMixed2,587 tokens in use

dbs-human-dispatch

Hands a task to an employee or collaborator with a written brief, follows replies and progress, and checks the delivered work against fixed criteria.

11,518 stars on the repository Custom / Unknowndontbesilent2025/dbskill · skills/dbs-human-dispatch
npx -y skills add dontbesilent2025/dbskill -g --all
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing dbs-human-dispatch costs

135tokens idle
2,587tokens in use
6,739tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Mixed

Script-ledMixedModel-led

Scripts do part of the work and the model's judgement does the rest.

This describes the skill's structure, how much of it is scripts. It does not say how well it works. Scripts: scripts/dispatch.py, scripts/test_dispatch.py.

Touches

shelldetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter8/15
  • Structure16/25
  • Examples0/15
  • Output0/15
  • Size15/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Runs shell commands−5

    does this on the machine, and the frontmatter does not say so

    scripts/dispatch.py, line 156

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community64
  • Maturity28
  • Momentum40

dbs-human-dispatch helps a boss delegate work to people. It reads the staff profiles and existing tasks in a folder you name, suggests who fits with evidence and unknowns, and writes a brief with the purpose, deliverables, scope, acceptance criteria, deadline with its basis, materials and reply method.

With the Feishu command line tool and your authorization it can create the task and read replies. Without them it hands you text ready to forward. After delivery it checks each criterion with evidence and gives a verdict of passed, needs revision or needs the boss to decide.

What You Get

  • A shortlist of candidate owners with the evidence for each and what is still unknown
  • A task brief from a template with deliverables, acceptance criteria, deadline and reply instructions
  • A forwardable message when no Feishu tool or permission is available
  • A standard-by-evidence-by-verdict acceptance table and a report for the boss
  • A local task record with state history, and optionally an event queue and agent worker configuration for automatic follow-up

Common Use Cases

  • Assigning a content or research task to a team member with clear acceptance criteria
  • Answering a staff member’s question or blocker within the authority you granted
  • Checking submitted work item by item before telling the boss it is done

What Makes It Unique

It separates a message being sent from a person accepting, and a submission from an accepted result. States change only on evidence such as a returned task ID. Staff can decline, negotiate the deadline or report a blocker. The skill makes no overtime plans, builds no rankings and treats text in staff documents as data. It falls back to a forwardable message whenever tools or authorization are missing, and never claims that an employee finished or a boss was notified without proof.

Under The Hood

It loads SKILL.md and three reference files, plus brief and profile templates. A bundled Python script, dispatch.py, keeps tasks, history, events and jobs in a local SQLite database under a dot folder in your home directory. With a Feishu CLI such as lark-cli installed and authorized, the agent runs it to create tasks, read documents, subscribe to document edit events, listen for events and send replies over the network. An optional worker command you configure is run by the script as a subprocess to process queued edits, only when you pass the allow flag. The agent reads local staff files you point it to. It writes task records and configuration files.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers