CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

linear

Search, read, create, and update the user's Linear issues, projects, and comments through their own Linear connection.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Linear

Use this skill when the user asks about Linear issues, tickets, projects, cycles, or their team's work tracking — finding issues, checking status, filing a bug, leaving a comment, or updating an issue.

Linear's API is GraphQL at https://api.linear.app/graphql, reached one of two ways:

  • Direct token: $VAULT_TOKEN_API_LINEAR_APP holds the user's Linear token; pass it as -H "Authorization: Bearer $VAULT_TOKEN_API_LINEAR_APP". The examples below use it as written.

  • App connection: with no token, the user connects Linear as an app instead (composio skill). Send the same request through QM — same method, URL, query, headers and JSON body, but no Authorization header:

    curl -sS "$AGENT_API_URL/v1/composio/proxy" -H "X-Agent-Capability: $AGENT_API_TOKEN" \
      -H 'content-type: application/json' \
      --data '{"method":"POST","url":"https://api.linear.app/graphql","body":{"query":"{ viewer { name email } }"}}'

    The reply is the provider's own status and body; the composio skill's "Raw requests" section covers 409 answers, choosing an account, and file bodies. The GraphQL {"query":…} object goes in body.

Do not ask the user for a token, log it, or use another principal's credential. With no connection, or a 401/400 (authentication) from the API, tell the user they need to connect Linear and stop — don't guess at issue data.

Find / read issues

Query issues by text, assignee, or state:

curl -sS -X POST 'https://api.linear.app/graphql' \
  -H "Authorization: Bearer $VAULT_TOKEN_API_LINEAR_APP" \
  -H 'content-type: application/json' \
  --data '{"query":"{ issues(first: 20, filter: { title: { containsIgnoreCase: \"login\" } }) { nodes { identifier title state { name } assignee { name } url } } }"}'

Read one issue by id/identifier:

curl -sS -X POST 'https://api.linear.app/graphql' \
  -H "Authorization: Bearer $VAULT_TOKEN_API_LINEAR_APP" \
  -H 'content-type: application/json' \
  --data '{"query":"{ issue(id: \"ENG-123\") { identifier title description state { name } assignee { name } url comments { nodes { body user { name } } } } }"}'

Keep issue identifiers and urls in your answer so claims can be traced.

To create or assign an issue you usually need a team id — list them once:

curl -sS -X POST 'https://api.linear.app/graphql' \
  -H "Authorization: Bearer $VAULT_TOKEN_API_LINEAR_APP" \
  -H 'content-type: application/json' \
  --data '{"query":"{ teams(first: 100) { nodes { id key name } pageInfo { hasNextPage endCursor } } }"}'

Linear paginates every list: if pageInfo.hasNextPage is true, re-query with after: "<endCursor>" — never conclude a team or issue doesn't exist from one page.

Writes require approval

Creating an issue, commenting, or changing state is a write. Prepare the exact mutation, summarize what it will create/change (team, title, target issue), and ask for approval before running it.

After approval, create an issue:

curl -sS -X POST 'https://api.linear.app/graphql' \
  -H "Authorization: Bearer $VAULT_TOKEN_API_LINEAR_APP" \
  -H 'content-type: application/json' \
  --data '{"query":"mutation($input: IssueCreateInput!) { issueCreate(input: $input) { success issue { identifier url } } }","variables":{"input":{"teamId":"TEAM_ID","title":"Login button misaligned on mobile","description":"Reported by a user; repro on iOS Safari."}}}'

Add a comment to an existing issue:

curl -sS -X POST 'https://api.linear.app/graphql' \
  -H "Authorization: Bearer $VAULT_TOKEN_API_LINEAR_APP" \
  -H 'content-type: application/json' \
  --data '{"query":"mutation($input: CommentCreateInput!) { commentCreate(input: $input) { success comment { url } } }","variables":{"input":{"issueId":"ISSUE_UUID","body":"Fixed in the latest deploy."}}}'

Always report the resulting issue/comment identifier and url.

Repository

yc-software/qm

Path

skills-seed/linear/SKILL.md

Last updated
First committed

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.