Install
$ agentstack add skill-alexandrbasis-claudops-cc-linear ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ● Network access Used
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
CC Linear
> Prefix the first response in a session with a one-line tag: [cc-linear].
Interact with Linear via .claude/scripts/linear-api.sh — a direct curl wrapper over Linear's GraphQL API. Each command is atomic and deterministic.
Commands
Issues
.claude/scripts/linear-api.sh get-issue TEAM-66
.claude/scripts/linear-api.sh search "authentication" 5
.claude/scripts/linear-api.sh ai-search "error handling in mobile app" 5
.claude/scripts/linear-api.sh create-issue "Title" "Description" 3
.claude/scripts/linear-api.sh create-issue "Title" "-" 3 <<< "Multiline desc"
.claude/scripts/linear-api.sh update-issue TEAM-66 --title "New Title"
.claude/scripts/linear-api.sh update-issue TEAM-66 --description "New desc"
.claude/scripts/linear-api.sh update-issue TEAM-66 --priority 2
.claude/scripts/linear-api.sh update-status TEAM-66 "In Progress"
.claude/scripts/linear-api.sh add-comment TEAM-66 "Comment body"
.claude/scripts/linear-api.sh add-comment TEAM-66 "-" <<'EOF'
Multiline comment here
EOF
.claude/scripts/linear-api.sh list-comments TEAM-66
Use "-" as the body argument to read from stdin — preferred for any description or comment longer than one line or containing quotes.
Labels, Assignment & Relations
.claude/scripts/linear-api.sh add-label TEAM-66 "Bug"
.claude/scripts/linear-api.sh remove-label TEAM-66 "Bug"
.claude/scripts/linear-api.sh assign TEAM-66 "Alexander Basis"
.claude/scripts/linear-api.sh add-relation TEAM-66 TEAM-67 "blocks"
.claude/scripts/linear-api.sh link-pr TEAM-66 "https://github.com/org/repo/pull/123"
Listings & Queries
.claude/scripts/linear-api.sh list-issues 10
.claude/scripts/linear-api.sh my-issues # Current user's active issues
.claude/scripts/linear-api.sh my-issues 20 # With custom limit
.claude/scripts/linear-api.sh list-states
.claude/scripts/linear-api.sh list-labels
.claude/scripts/linear-api.sh list-users
Output Formatting
The script returns raw JSON. Present results to the user clearly:
- get-issue: Show identifier, title, status, priority, assignee, labels, and URL
- search / ai-search: Numbered list with identifier, title, and status
- create-issue: Confirm with identifier and URL
- update-status / update-issue: Confirm the change (e.g., "TEAM-66: Todo → In Progress")
- list-comments: Show author, date, and body for each comment
- my-issues: Show as a compact table grouped by status
- list-*: Show as a compact table
Include the Linear URL in every issue response — the user will want to click through.
Before Acting
Linear mutations are visible to the whole team and send notifications. Before running any of these commands, echo back what you are about to do and wait for confirmation on first use in a session:
create-issue,add-comment,add-relation,link-pr(new visible content)update-status,update-issue,assign,add-label,remove-label(state changes)
Read-only commands (get-issue, search, ai-search, list-*, my-issues) do not need confirmation — run them freely.
For multi-step patterns below, confirm the whole chain once, then execute without re-prompting between steps.
Configuration
| Setting | Value | |---------|-------| | Team Key | LINEAR_TEAM_KEY or default TEAM | | API Key | LINEAR_API_KEY env var |
Task States (in order)
Backlog → Todo → In Progress → In Review → Done | Canceled | Duplicate
Priority Levels
0: None 1: Urgent 2: High 3: Normal (default) 4: Low
Workflow Patterns
Common multi-step sequences. Each command is atomic. When the steps have no dependency on each other (e.g. adding a label and assigning at the same time), batch them in a single turn; run sequentially only when a later step needs output from an earlier one.
Start Implementation
.claude/scripts/linear-api.sh update-status TEAM-66 "In Progress"
.claude/scripts/linear-api.sh add-comment TEAM-66 "Implementation started. Branch: feature/team-66-feature-name"
Submit for Review
.claude/scripts/linear-api.sh update-status TEAM-66 "In Review"
.claude/scripts/linear-api.sh add-comment TEAM-66 "-" <<'EOF'
Implementation completed.
- Key changes: [list]
- Test coverage: [X]%
- PR ready for review.
EOF
Task Done
.claude/scripts/linear-api.sh update-status TEAM-66 "Done"
.claude/scripts/linear-api.sh add-comment TEAM-66 "Task completed and PR merged. SHA: abc123"
Cross-Skill Integration
When other skills touch Linear-adjacent work, offer the matching Linear operation as a suggestion — do not run it unprompted. Only execute when the user agrees or when the outer skill explicitly delegates to you.
| Skill | When | Linear Action | |-------|------|---------------| | /ct | After creating task docs | create-issue with task title + link to task doc | | /si | On implementation start | update-status → "In Progress" + comment with branch | | /sr | After code review passes | update-status → "In Review" or "Done" | | PR creation | After gh pr create | link-pr + update-status → "In Review" |
Don't force these — suggest them when the context is clear (e.g., a TEAM-* ID is visible in the task doc or branch name).
Error Recovery
- If a command fails mid-sequence, retry the failed command — completed steps are idempotent.
- If "State not found", run
list-statesto refresh the cache, then retry. - If "Label/User not found", run
list-labelsorlist-usersto check exact names. - Clear all caches:
rm /tmp/linear-TEAM-*.json - Report command failures (non-zero exits, GraphQL errors) to the user with the
exact error message — don't retry hidden.
References
See references/linear-api-reference.md for GraphQL queries used by the script.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: alexandrbasis
- Source: alexandrbasis/claudops
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.