agentleFS
OpenCovenOpenCoven/coven/skills/linear-issue-management/SKILL.md

linear-issue-management

Skill75 starsChanged 4 months agoMITView on GitHub

Manage Linear issues safely through the GraphQL API, including team/project/label discovery, creating issues, updating issues, searching existing issues, handling mutually-exclusive labels, and using 1Password-backed Linear API auth without printing secrets. Use when filing, triaging, editing, or auditing Linear tickets such as CORE issues.

  • Reads credentials

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

19 rules. About 0 are common, 14 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈14

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when working with Linear issues through the API.

    line 8
  2. Distinctive:

    Search for an existing issue to avoid duplicates.

    line 18
  3. Distinctive:

    Do not print, paste, log, or commit the Linear API token.

    line 25
  4. Distinctive:

    Prefer 1Password-backed env injection or an existing secret manager.

    line 26
  5. Distinctive:

    Treat issue creation, updates, comments, assignment changes, and state transitions as external state-changing actions.

    line 27
  6. Distinctive:

    Use GraphQL variables. Do not interpolate untrusted user text into a GraphQL query string.

    line 28
  7. Distinctive:

    Keep issue text factual. Do not file speculation as confirmed defects.

    line 29
  8. Distinctive:

    Do not file in a team just because an issue key prefix looks familiar.

    line 30
  9. Very specific:

    Use a Linear API key in LINEAR_API_KEY.

    line 34
  10. Very specific:

    Never echo the token. If debugging auth, print only whether LINEAR_API_KEY is set, not its value.

    line 51
  11. Very specific:

    Read references/graphql-recipes.md for query/mutation templates.

    line 55
  12. Distinctive:

    Search existing issues by relevant keywords and repo/project names.

    line 57
  13. Very specific:

    Create the issue with title, description, teamId, optional projectId, optional labelIds, and optional priority.

    line 60
  14. Distinctive:

    If two requested labels conflict, choose the more precise/current one and mention the omitted label.

    line 66
  15. Distinctive:

    Use this compact structure unless the user asks otherwise:

    line 71
  16. Very specific:

    Search by issue key first if provided (CORE-14).

    line 95
  17. Distinctive:

    Search by URL slug or title keywords when no key is available.

    line 96
  18. Distinctive:

    Prefer additive updates unless the user explicitly asks to remove/replace labels, state, assignee, or project.

    line 98
  19. Distinctive:

    When handing a Linear issue to Cody or another coding agent, include:

    line 102

≈ “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

Linear Issue Management

Use this skill when working with Linear issues through the API.

Prime directive

Discover before writing. Never guess the team, project, state, assignee, or label IDs from memory when creating or updating a Linear issue.

Before any mutation:

  1. Confirm the target workspace/team from the user request or current project context.
  2. Query Linear for the exact team/project/label/state IDs.
  3. Search for an existing issue to avoid duplicates.
  4. Show the intended mutation summary if the action is externally visible or ambiguous.
  5. Execute one focused GraphQL mutation.
  6. Return the issue key + URL and record any important follow-up context.

Safety rules

  • Do not print, paste, log, or commit the Linear API token.
  • Prefer 1Password-backed env injection or an existing secret manager. Do not write tokens to files.
  • Treat issue creation, updates, comments, assignment changes, and state transitions as external state-changing actions. Get explicit user approval unless the user already gave a direct instruction in the current conversation.
  • Use GraphQL variables. Do not interpolate untrusted user text into a GraphQL query string.
  • Keep issue text factual. Do not file speculation as confirmed defects.
  • Do not file in a team just because an issue key prefix looks familiar; confirm via API.

Auth pattern

Use a Linear API key in LINEAR_API_KEY.

Preferred shape:

op run --env-file <linear-env-file> -- <command using LINEAR_API_KEY>

or, when the shell already has the token:

curl https://api.linear.app/graphql \
  -H "Authorization: $LINEAR_API_KEY" \
  -H 'Content-Type: application/json' \
  --data-binary @payload.json

Never echo the token. If debugging auth, print only whether LINEAR_API_KEY is set, not its value.

Workflow: create an issue

  1. Read references/graphql-recipes.md for query/mutation templates.
  2. Discover teams and choose the exact teamId.
  3. Search existing issues by relevant keywords and repo/project names.
  4. Discover labels/projects/states only after the team is known.
  5. Resolve label conflicts before mutation.
  6. Create the issue with title, description, teamId, optional projectId, optional labelIds, and optional priority.
  7. Return the created issue identifier and URL.

Label rules

  • Linear can have grouped labels where child labels are mutually exclusive.
  • If two requested labels conflict, choose the more precise/current one and mention the omitted label.
  • Known OpenClaw Linear rule: Infra and Improvement are mutually exclusive grouped child labels. Do not apply both.

Writing good issue bodies

Use this compact structure unless the user asks otherwise:

## Problem
<what is wrong and who it affects>

## Evidence
- <specific file/log/error/link when available>

## Expected behavior
<desired end state>

## Proposed fix
<short implementation direction, if known>

## Acceptance criteria
- [ ] <verifiable outcome>
- [ ] <test or validation gate>

For follow-up implementation tickets, include exact repo paths, config keys, issue IDs, and commands when known.

Updating/searching issues

  • Search by issue key first if provided (CORE-14).
  • Search by URL slug or title keywords when no key is available.
  • Fetch current labels/state before replacing anything.
  • Prefer additive updates unless the user explicitly asks to remove/replace labels, state, assignee, or project.

Handoff to coding agents

When handing a Linear issue to Cody or another coding agent, include:

  • Linear key + URL
  • Repo/path scope
  • Summary of the intended code change
  • Required verification commands
  • Constraints: preserve dirty work, no secret printing, do not touch unrelated files
  • Any linked docs/memory references needed for context

References

  • references/graphql-recipes.md — GraphQL discovery, search, create, update, and comment templates.

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.