agentleFS
Sign inStart free
LIDR-academyLIDR-academy/AI4Devs-LTI-extended/.claude/skills/openspec-ff-change/SKILL.md

openspec-ff-change

Skill278 starsChanged 6 months agoMITView on GitHub

Fast-forward through OpenSpec artifact creation. Use when the user wants to quickly create all artifacts needed for implementation without stepping through each one individually.

We split this skill into the rules it gives an agent

41 rules. 11 are common, 14 are distinctive, and 16 are very specific to this project.

  • 11

    Common

    Most of the 59 edited versions we read keep these.

  • 14

    Distinctive

    This author's own choices. Worth a look.

  • 16

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use getAccessibleAtlassianResources MCP tool to get the cloudId

    line 24
  2. Very specific:

    Use getJiraIssue MCP tool with:

    line 25
  3. Distinctive:

    Remove any leading/trailing hyphens

    line 32
  4. Distinctive:

    If ticket title is unclear or too long, use a shortened meaningful version

    line 34
  5. Very specific:

    Use the derived kebab-case name as <name> for the change directory

    line 35
  6. Distinctive:

    Use ticket content as context for creating artifacts

    line 36
  7. Distinctive:

    Use the provided name directly

    line 40
  8. Distinctive:

    Check if change already exists, if so ask user if they want to continue it

    line 41
  9. Common:

    Use the AskUserQuestion tool (open-ended, no preset options) to ask:

    line 47
  10. Common:

    IMPORTANT: Do NOT proceed without understanding what the user wants to build.

    line 50
  11. Distinctive:

    If the user has attached files to this conversation:

    line 59
  12. Distinctive:

    Check for any files in the conversation context (attached files will be visible in the file list)

    line 60
  13. Distinctive:

    Read the file to get its current path

    line 62
  14. Distinctive:

    Use the file system tools to copy/move the file, preserving the original filename

    line 64
  15. Distinctive:

    If files were moved, inform the user: "Moved N attached file(s) to the change directory root."

    line 65
  16. Common:

    Use the TodoWrite tool to track progress through the artifacts.

    line 77
  17. Very specific:

    CRITICAL for tasks artifact: If creating tasks.md:

    line 93
  18. Very specific:

    Read openspec/config.yaml to get backend-specific rules (mandatory steps, branch naming, etc.)

    line 94
  19. Very specific:

    Read .claude/rules/openspec-tasks-mandatory-steps.mdc to understand mandatory testing requirements and agent execution responsibilities

    line 95
  20. Distinctive:

    All mandatory steps that MUST be included (e.g., Step 0: Create Feature Branch)

    line 97
  21. Distinctive:

    If Jira ticket was provided: Use ticket content to inform artifact creation (especially proposal and tasks)

    line 98
  22. Common:

    Read any completed dependency files for context

    line 99
  23. Very specific:

    Create the artifact file using template as the structure

    line 100
  24. Very specific:

    For tasks artifact: Ensure all mandatory steps from config.yaml and the rule file are included:

    line 102
  25. Distinctive:

    Run Unit Tests and Verify Database State (MANDATORY)

    line 105
  26. Distinctive:

    Update Technical Documentation (MANDATORY)

    line 108
  27. Very specific:

    After creating each artifact, re-run openspec status --change "<name>" --json

    line 113
  28. Very specific:

    Check if every artifact ID in applyRequires has status: "done" in the artifacts array

    line 114
  29. Very specific:

    Stop when all applyRequires artifacts are done

    line 115
  30. Common:

    Use AskUserQuestion tool to clarify

    line 118
  31. Common:

    After completing all artifacts, summarize:

    line 128
  32. Very specific:

    Follow the instruction field from openspec instructions for each artifact type

    line 136
  33. Common:

    Read dependency artifacts for context before creating new ones

    line 138
  34. Very specific:

    Use template as the structure for your output file - fill in its sections

    line 139
  35. Very specific:

    Do NOT copy <context>, <rules>, <project_context> blocks into the artifact

    line 141
  36. Very specific:

    Create ALL artifacts needed for implementation (as defined by schema's apply.requires)

    line 145
  37. Common:

    Always read dependency artifacts before creating a new one

    line 146
  38. Very specific:

    For tasks.md: Read .claude/rules/openspec-tasks-mandatory-steps.mdc to ensure all mandatory steps are included with proper agent execution …

    line 147
  39. Common:

    If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum

    line 148
  40. Common:

    If a change with that name already exists, suggest continuing that change instead

    line 149
  41. Common:

    Verify each artifact file exists after writing before proceeding to next

    line 150

“Common” is measured: kept by at least 6 in 10 of the 59 edited versions of this file we could read. “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

Fast-forward through artifact creation - generate everything needed to start implementation in one go.

Input: The user's request should include:

  • A Jira ticket ID (e.g., SCRUM-123) - will fetch ticket content using Jira MCP
  • A change name (kebab-case) - will use that name directly
  • A description of what they want to build - will derive a kebab-case name

Steps

  1. Determine input type and get context

    a. If input looks like a Jira ticket ID (matches pattern like SCRUM-123, PROJ-456, etc.):

    • Use getAccessibleAtlassianResources MCP tool to get the cloudId
    • Use getJiraIssue MCP tool with:
      • cloudId: from step above
      • issueIdOrKey: the provided ticket ID
    • Extract ticket content (title, description, acceptance criteria, etc.)
    • Derive a kebab-case change name from the ticket title:
      • Convert ticket title to lowercase
      • Replace spaces and special characters with hyphens
      • Remove any leading/trailing hyphens
      • Example: "Update Position API" → update-position-api, "Add User Auth" → add-user-auth
      • If ticket title is unclear or too long, use a shortened meaningful version
    • Use the derived kebab-case name as <name> for the change directory
    • Use ticket content as context for creating artifacts
    • Store ticket ID for reference (e.g., in proposal or as metadata)

    b. If input is a change name (kebab-case format):

    • Use the provided name directly
    • Check if change already exists, if so ask user if they want to continue it

    c. If input is a description:

    • Derive a kebab-case name (e.g., "add user authentication" → add-user-auth)

    d. If no input provided:

    • Use the AskUserQuestion tool (open-ended, no preset options) to ask:

      "What change do you want to work on? Provide a Jira ticket ID (e.g., SCRUM-123), change name, or describe what you want to build."

    IMPORTANT: Do NOT proceed without understanding what the user wants to build.

  2. Create the change directory
    openspec new change "<name>"
    This creates a scaffolded change at openspec/changes/<name>/.

2.5. Handle attached files (if any) If the user has attached files to this conversation:

  • Check for any files in the conversation context (attached files will be visible in the file list)
  • For each attached file:
    • Read the file to get its current path
    • Move it to the root of the change directory: openspec/changes/<name>/<filename>
    • Use the file system tools to copy/move the file, preserving the original filename
  • If files were moved, inform the user: "Moved N attached file(s) to the change directory root."
  1. Get the artifact build order
    openspec status --change "<name>" --json
    Parse the JSON to get:
    • applyRequires: array of artifact IDs needed before implementation (e.g., ["tasks"])
    • artifacts: list of all artifacts with their status and dependencies

  2. Create artifacts in sequence until apply-ready

    Use the TodoWrite tool to track progress through the artifacts.

    Loop through artifacts in dependency order (artifacts with no pending dependencies first):

    a. For each artifact that is ready (dependencies satisfied):

    • Get instructions:
      openspec instructions <artifact-id> --change "<name>" --json
    • The instructions JSON includes:
      • context: Project background (constraints for you - do NOT include in output)
      • rules: Artifact-specific rules (constraints for you - do NOT include in output)
      • template: The structure to use for your output file
      • instruction: Schema-specific guidance for this artifact type
      • outputPath: Where to write the artifact
      • dependencies: Completed artifacts to read for context
    • CRITICAL for tasks artifact: If creating tasks.md:
      • Read openspec/config.yaml to get backend-specific rules (mandatory steps, branch naming, etc.)
      • Read .claude/rules/openspec-tasks-mandatory-steps.mdc to understand mandatory testing requirements and agent execution responsibilities
      • Task structure requirements
      • All mandatory steps that MUST be included (e.g., Step 0: Create Feature Branch)
    • If Jira ticket was provided: Use ticket content to inform artifact creation (especially proposal and tasks)
    • Read any completed dependency files for context
    • Create the artifact file using template as the structure
    • Apply context and rules as constraints - but do NOT copy them into the file
    • For tasks artifact: Ensure all mandatory steps from config.yaml and the rule file are included:
      • Step 0: Create Feature Branch (MUST be first step for backend changes)
      • Review and Update Existing Unit Tests (MANDATORY)
      • Run Unit Tests and Verify Database State (MANDATORY)
      • Manual Endpoint Testing with curl (MANDATORY - AGENT MUST EXECUTE)
      • E2E Testing with Playwright MCP (MANDATORY if applicable - AGENT MUST EXECUTE)
      • Update Technical Documentation (MANDATORY)
    • For manual testing tasks: Include sub-tasks that make it clear the agent must execute tests (e.g., "Test GET endpoints with curl", "Restore database state", etc.)
    • Show brief progress: "✓ Created <artifact-id>"

    b. Continue until all applyRequires artifacts are complete

    • After creating each artifact, re-run openspec status --change "<name>" --json
    • Check if every artifact ID in applyRequires has status: "done" in the artifacts array
    • Stop when all applyRequires artifacts are done

    c. If an artifact requires user input (unclear context):

    • Use AskUserQuestion tool to clarify
    • Then continue with creation
  3. Show final status
    openspec status --change "<name>"

Output

After completing all artifacts, summarize:

  • Change name and location
  • List of artifacts created with brief descriptions
  • What's ready: "All artifacts created! Ready for implementation."
  • Prompt: "Run /opsx:apply or ask me to implement to start working on the tasks."

Artifact Creation Guidelines

  • Follow the instruction field from openspec instructions for each artifact type
  • The schema defines what each artifact should contain - follow it
  • Read dependency artifacts for context before creating new ones
  • Use template as the structure for your output file - fill in its sections
  • IMPORTANT: context and rules are constraints for YOU, not content for the file
    • Do NOT copy <context>, <rules>, <project_context> blocks into the artifact
    • These guide what you write, but should never appear in the output

Guardrails

  • Create ALL artifacts needed for implementation (as defined by schema's apply.requires)
  • Always read dependency artifacts before creating a new one
  • For tasks.md: Read .claude/rules/openspec-tasks-mandatory-steps.mdc to ensure all mandatory steps are included with proper agent execution requirements
  • If context is critically unclear, ask the user - but prefer making reasonable decisions to keep momentum
  • If a change with that name already exists, suggest continuing that change instead
  • Verify each artifact file exists after writing before proceeding to next

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.

Reviews can't be read right now.

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.

Check your setup, free