LI

linear-symphony-intake

Routes and shapes yandex.ad tasks into the Linear-Symphony pipeline.

Install

mkdir -p .claude/skills/linear-symphony-intake && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/13017" && unzip -o skill.zip -d .claude/skills/linear-symphony-intake && rm skill.zip

Installs to .claude/skills/linear-symphony-intake

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Use when the user wants to create, refine, split, or update Linear issues for yandex.ad so Symphony can execute them safely. Covers issue shaping, ownership boundaries, release-required routing, client handoff vs client-repo changes, and label selection for the feature -> PR -> release pipeline.
296 chars✓ has a “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Intermediate

Key capabilities

  • Create or refine Linear issues for `yandex.ad`
  • Shape issues for Symphony execution
  • Determine issue class (bug, feature, investigation, release)
  • Assess risk (low, medium, high)
  • Define required capabilities and external inputs

How it works

The skill guides the user through required intake decisions, applies label rules, and then drafts or refines the issue body before previewing or creating the Linear issue.

Inputs & outputs

You give it
User request to create, refine, split, or update Linear issues
You get back
A Linear issue ready for a human to move from `Backlog` to `Todo`

When to use linear-symphony-intake

  • Creating new feature issues
  • Refining bug reports for Symphony
  • Routing tasks for release
  • Managing downstream dependencies

About this skill

Linear Symphony Intake

Use this skill for new yandex.ad work that should enter the Linear + Symphony pipeline.

Read these files before shaping the issue:

  1. docs/automation/issue-writing-rules.md
  2. docs/automation/linear-intake.md
  3. references/checklist.md

Choose the smallest fitting template:

  • feature: docs/automation/templates/linear-feature.md
  • bug: docs/automation/templates/linear-bug.md
  • investigation: docs/automation/templates/linear-investigation.md
  • release: docs/automation/templates/linear-release.md
  • client workflow / downstream dependency: docs/automation/templates/linear-marketing2025-workflow.md

Required intake decisions

Before creating or updating an issue, capture or infer:

  1. issue class: bug / feature / investigation / release
  2. risk: low / medium / high
  3. owner repo
  4. out-of-scope repos
  5. Release Required: yes/no
  6. client handoff required: yes/no
  7. compatibility task vs migration task
  8. required capabilities:
    • browser mode
    • live-api yes/no
    • manual-check yes/no
    • operator step required yes/no
  9. external inputs / secrets:
    • required env vars
    • source of truth
    • expected parent-process availability
  10. blocked input policy
  11. acceptance criteria
  12. Feature Validation
  13. PR Validation
  14. Release Validation

If the task is ambiguous, default to the smaller producer-side issue and create a downstream follow-up issue later.

Label rules

For a new Symphony feature issue, default to:

  • symphony
  • issue-type:feature

Add:

  • domain labels such as search-api, wordstat, dashboard
  • release-required only when the client cannot use the work before a published image exists

Do not create issue-type:pr or issue-type:release manually during normal intake. Those are generated as follow-up issues by the pipeline.

Workflow

  1. Draft or refine the Markdown issue body.
  2. Check that repo ownership and handoff boundaries are explicit.
  3. Ensure capability requirements and secret dependencies are explicit before the issue can move to Todo.
  4. Ensure validation is defined per stage, not as one mixed checklist.
  5. Ensure the issue does not require client-repo edits unless that repo is explicitly in scope.
  6. Preview the Linear payload with python scripts/linear_issue.py preview.
  7. Create or update the Linear issue only when the user asks for the actual Linear change.

Commands

Preview:

python scripts/linear_issue.py preview \
  --config /Users/georgyagaev/Projects/Symphony_yaad/linear.yandexad.json \
  --from <draft.md> \
  --title "<issue title>" \
  --labels symphony,issue-type:feature,<domain-label>

Create:

python scripts/linear_issue.py create \
  --config /Users/georgyagaev/Projects/Symphony_yaad/linear.yandexad.json \
  --from <draft.md> \
  --title "<issue title>" \
  --labels symphony,issue-type:feature,<domain-label>

Update:

python scripts/linear_issue.py update \
  --config /Users/georgyagaev/Projects/Symphony_yaad/linear.yandexad.json \
  --issue-id GEO-7 \
  --from <draft.md> \
  --title "<issue title>"

Output standard

The finished issue should be ready for a human to move from Backlog to Todo without extra copy-paste or manual rewriting.

When not to use it

  • When the work is not for `yandex.ad`
  • When the work should not enter the `Linear + Symphony` pipeline
  • When creating `issue-type:pr` or `issue-type:release` manually during normal intake

Limitations

  • Does not create `issue-type:pr` or `issue-type:release` manually
  • Defaults to smaller producer-side issue if task is ambiguous
  • Does not require client-repo edits unless that repo is explicitly in scope

How it compares

This skill structures the creation of Linear issues with specific rules and validations for `yandex.ad` and Symphony, unlike generic issue tracking.

Compared to similar skills

linear-symphony-intake side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
linear-symphony-intake (this skill)01moReviewIntermediate
github-project-management46moReviewAdvanced
create-plans18moReviewIntermediate
phasing17moNo flagsIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry