Agent skill
linear-crit-path-to-task
{{ 𝛀𝛀𝛀 }} Decompose a Linear issue into a critical path of granular issues
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/linear-crit-path-to-task
SKILL.md
Objective
Decompose Linear issue FOU-$ARGUMENTS into a set of granular, dependency-ordered sub-issues that form a critical path from the current state of the codebase to the completed feature.
Process
1. Gather Context
- Fetch the full details of
FOU-$ARGUMENTSfrom Linear (include relations) - Analyse the codebase to understand the current state relevant to this issue
- List all active Linear projects to understand where issues should be placed
2. Decompose
Break the issue into the smallest meaningful units of work. Each issue should be:
- Independently deliverable — can be completed, reviewed, and merged on its own
- Single responsibility — one clear outcome per issue
- Well-scoped — achievable in a focused session (hours, not days)
3. Classify Relationships
Apply consistent reasoning to distinguish between two relationship types:
Child issues (use parentId) when the work is:
- A direct sub-task that exists solely to accomplish part of the parent
- Meaningless without the parent issue's context
- Something that would naturally nest under the parent in a work breakdown
Blocking issues (use blocks/blockedBy) when the work is:
- A prerequisite that has independent value beyond the parent issue
- Infrastructure, refactoring, or setup that other work also depends on
- Something that must be completed first but isn't conceptually "part of" the parent
An issue can be both a child AND have blocking relationships with siblings.
4. Determine Project Placement
For each issue, select the most appropriate existing Linear project based on:
- The issue's domain (frontend, backend, infrastructure, etc.)
- Which project's scope best encompasses the work
- If no project fits well, omit the project field rather than forcing a bad fit
5. Build the Critical Path
Order the issues by dependency. The critical path should flow logically:
- Foundation/infrastructure work first
- Core implementation in the middle
- Integration, polish, and verification last
6. Present the Plan
Before creating anything, present a summary table showing:
- Issue title
- Relationship type (child / blocker)
- Dependencies (what it blocks / is blocked by)
- Target project
- Priority (2=High, 3=Normal, 4=Low)
Wait for approval before creating issues.
7. Create Issues
Once approved, create all issues in dependency order (foundations first). For each issue:
- Assign to
me - Set the appropriate
parentIdfor child issues - Set
blocks/blockedByusing the identifiers of previously created issues - Set
projectif determined in step 4 - Set
prioritybased on critical path position - Write a clear description in markdown including:
- What needs to be done
- Acceptance criteria
- Any relevant context about the current codebase state
8. Confirm
After all issues are created, provide a summary of what was created with their identifiers and a visual representation of the dependency graph.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?