Skip to content
SkillModel-led2,382 tokens in use

Flight Deck

A status board for long agent work: STATUS.md, LOG.md and an owner page that survive compaction and show only what needs you.

680 stars on the repository MIT Licensephuryn/product-engineering · agent-harness/skills/flight-deck
npx skills add phuryn/product-engineering
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Flight Deck costs

42tokens idle
2,382tokens in use
15,590tokens, 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 Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

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 →
Good
  • Frontmatter8/15
  • Structure19/25
  • Examples3/15
  • Output15/15
  • Size15/15
  • Resources15/15

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

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity100
  • Maintenance100
  • Community44
  • Maturity4
  • Momentum40

Flight Deck is for work that outlives a context window: multi-hour runs, background jobs, waiting on a decision, or any request like “where were we” or a status report. It keeps state in files that cannot be summarized away.

STATUS.md is rewritten in place to say only what is true now, LOG.md records decisions and dead ends, and an owner page shows a short status plus the decisions, tasks and gates that need a person.

What You Get

  • A STATUS.md that is rewritten on every state change and read first after a compaction
  • An append-only LOG.md of decisions, corrections and dead ends
  • An owner page template with phase, milestones, a board of key work items and answerable asks
  • Rules for consuming answers: record the decision, delete the item, then act
  • A OWNER.md fallback for agents without the Artifact tool

Common Use Cases

  • Running a multi-hour build with background jobs
  • Answering a status question or producing a progress deck from one source of truth
  • Collecting decisions from an owner who is not watching the session
  • Resuming work after a context compaction without re-running paid jobs

What Makes It Unique

Status is rewritten in place, never appended, so every line is current, and evidence is recorded with its actual numbers so paid experiments are not repeated. Owner asks carry everything needed to answer them, including a recommendation, and a go on a one-way door must be typed in chat. A scheduled pass with no chat can consume answers but never passes a gate or publishes a page.

Under The Hood

The skill is instructions plus an HTML owner-page template and reference files. The agent writes STATUS.md and LOG.md in the repository, checks that listed processes are really running, and when the Artifact tool exists it publishes the page template and writes status and items to its database, then reads the saved answers. A reference describes scheduling a headless Claude Code pass with a shell command that reads a prompt from a file. The skill does not tell the agent to fetch general web pages.

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