agentleFS
nonlinear-xyznonlinear-xyz/factory-kit/skills/factory-entry/SKILL.md

factory-entry

Skill9 starsChanged 4 months agoMITView on GitHub

Load a Linear issue into context and enter plan mode for a focused session

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read project configuration: Prefer .factory-kit/linear.json.

    line 13
  2. Very specific:

    Resolve the identifier: If the invocation input is numeric, construct <teamKey>-<num> (e.g., 45 → NON-45).

    line 15
  3. Very specific:

    Check blockers: If the issue has open blockedBy relations, list them and warn loudly before proceeding.

    line 21
  4. Distinctive:

    Summarize for context (≤ 250 words): Use this shape — bold labels, short lines, omit any section that's empty:

    line 23
  5. Distinctive:

    Customer voice: if user feedback is quoted in the issue or linked, surface the raw words — don't paraphrase

    line 28
  6. Distinctive:

    Enter planning mode: Use the host's planning capability when available.

    line 34

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

You're starting a focused work session on a single Linear issue. Load its context, summarize it, and enter plan mode so the user can sign off before you start coding.

Invocation input: the issue identifier. Accept either the full form (NON-45) or just the number (45). If numeric, prepend the team key from the project configuration.

What to do

  1. Read project configuration. Prefer .factory-kit/linear.json; use legacy .claude/linear.json only as a migration fallback. If neither exists, stop and direct the user to the setup-linear workflow. Pull teamKey for ID resolution.
  2. Resolve the identifier. If the invocation input is numeric, construct <teamKey>-<num> (e.g., 45 → NON-45). If already in KEY-NN form, use it as-is.
  3. Fetch the issue through the connected Linear integration. Include comments and linked sub-issues if available.
  4. Fetch the project through the Linear integration if the issue belongs to one and you haven't loaded it yet this session.
  5. Check blockers. If the issue has open blockedBy relations, list them and warn loudly before proceeding. Ask whether the user wants to continue or pick a different ticket.
  6. Summarize for context (≤ 250 words). Use this shape — bold labels, short lines, omit any section that's empty:
    • Issue: <KEY-N> — title + priority + current state
    • Outcome: one sentence on what "done" observably looks like. If the issue lacks a clear outcome, say so — that's the first thing to fix.
    • Scope: the unchecked acceptance criteria / open checkboxes
    • Encoded constraints: anything in the description or comments that locks an approach — name the underlying force (compliance, perf budget, prior decision), not just the rule
    • Customer voice: if user feedback is quoted in the issue or linked, surface the raw words — don't paraphrase
    • Open design calls: "decide during build" items worth resolving up front
    • Context graph: parent, sub-issues, blockedBy/blocks, linked PRs, design docs — IDs + titles only. This is the issue's place in the wider decision graph; if it looks isolated, flag it as a yellow flag.
  7. Flag issue quality (brief — one line if everything is fine). Linear's house style is: short title with a concrete outcome, body only as long as needed. If the issue is missing a clear outcome, written as a user story, or sitting orphaned with no context links, name it. Don't fix it silently — surface it so the user can decide whether to rewrite before designing the change.
  8. Enter planning mode. Use the host's planning capability when available. Otherwise, present the implementation plan in chat and stop for explicit approval. Do not start coding before approval.

Style

Follow factory-voice.md. The reader is you, two minutes from now, about to design the change — give yourself what you'd need to make calls. Surface non-obvious constraints (invariants, prior decisions, blocking relations) at the top, not buried. No marketing copy, no hedging. If a constraint locks the design, name the principle behind it, not just the rule.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.