fstrent-task-management
Skill1 starsChanged 13 months agoMITView on GitHub
Manage project tasks using the fstrent_spec_tasks system. Use when creating, updating, tracking, or viewing tasks in .fstrent_spec_tasks/ folder. Handles task files, TASKS.md updates, status management, and task queries. Triggers on requests mentioning tasks, todos, work items, or task status.
In this repository40 other files it gives its agents.
- agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .cursor/rules/always.mdcCursor rule
- .cursor/rules/atlassian/atlassian_integration.mdcCursor rule
- .cursor/rules/computer_use/computer_use_agent.mdcCursor rule
- .cursor/rules/documentation.mdcCursor rule
- .cursor/rules/fstrent_spec_tasks/claude_code_cli.mdcCursor rule
- .cursor/rules/github/github_integration.mdcCursor rule
- .cursor/rules/powershell.mdcCursor rule
- .cursor/rules/silicon_valley_personality.mdcCursor rule
- .cursor/rules/virtual_environments_python.mdCursor rule
- .cursor/rules/web_tools/web_tools.mdcCursor rule
- artifacts-builderSkill
- atlassian-integrationSkill
- claude-code-cli-integrationSkill
- computer-use-agentSkill
- deep-researchSkill
- fstrent-code-reviewerSkill
- fstrent-planningSkill
- fstrent-qaSkill
- github-integrationSkill
- hanx-database-toolsSkill
- hanx-knowledge-baseSkill
- hanx-mediawikiSkill
- hanx-youtube-researcherSkill
- internal-commsSkill
- mcp-builderSkill
- oracle-apex-developmentSkill
- patent-filing-aiSkill
- Project SetupSkill
- research-methodologySkill
- skill-creatorSkill
- startup-business-formationSkill
- startup-product-developmentSkill
- startup-resource-accessSkill
- startup-vc-fundraisingSkill
- template-skillSkill
- web-toolsSkill
- youtube-rag-storageSkill
- youtube-video-analysisSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
49 rules. About 0 are common, 35 are distinctive, and 14 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈35
Distinctive · estimate
This author's own choices. Worth a look.
14
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Every task file begins with YAML frontmatter containing metadata:
- Distinctive: line 44
After the YAML frontmatter, include:
- Very specific: line 58
Create task file: .fstrent_spec_tasks/tasks/task{id}_descriptive_name.md
- Distinctive: line 59
Add YAML frontmatter with all required fields
- Distinctive: line 60
Add task content (objective, acceptance criteria, notes)
- Very specific: line 61
Update .fstrent_spec_tasks/TASKS.md with new entry
- Very specific: line 83
Update status field in task file's YAML frontmatter
- Very specific: line 84
Update corresponding entry in .fstrent_spec_tasks/TASKS.md
- Very specific: line 103
Read .fstrent_spec_tasks/TASKS.md for overview
- Very specific: line 112
Read .fstrent_spec_tasks/PROJECT_CONTEXT.md for goals and scope
- Distinctive: line 161
If a task is complex (affects multiple subsystems, >2-3 days effort, many acceptance criteria), break it into sub-tasks.
- Distinctive: line 189
Check dependency status before starting work
- Distinctive: line 190
Update dependent tasks when blockers are resolved
- Very specific: line 207
All tasks should align with project goals in PROJECT_CONTEXT.md.
- Distinctive: line 233
When working with tasks, automatically create missing folders and files:
- Distinctive: line 245
No confirmation needed: - Create files and folders silently, report what was created.
- Distinctive: line 250
Use clear, actionable titles
- Distinctive: line 257
Update status immediately when starting or completing
- Distinctive: line 258
Keep TASKS.md synchronized with task files
- Distinctive: line 259
Document blockers and issues
- Distinctive: line 260
Update acceptance criteria as work progresses
- Very specific: line 269
Only one task should be in-progress per developer at a time
- Very specific: line 271
Use failed status for blocked or abandoned work
- Distinctive: line 272
Document reason for failure in task notes
- Distinctive: line 278
Read TASKS.md to determine next ID
- Distinctive: line 279
Create task file with proper naming
- Distinctive: line 280
Add YAML frontmatter with all fields
- Distinctive: line 281
Add objective and acceptance criteria
- Distinctive: line 282
Update TASKS.md with new entry
- Distinctive: line 287
Read task005_*.md file
- Very specific: line 288
Update status: in-progress in YAML
- Very specific: line 289
Update TASKS.md entry from [ ] to [🔄]
- Distinctive: line 303
Verify acceptance criteria are met
- Distinctive: line 312
If TASKS.md or task files are missing:
- Very specific: line 313
Check if .fstrent_spec_tasks/ exists
- Distinctive: line 314
Create directory structure if needed
- Distinctive: line 319
If user references non-existent task:
- Distinctive: line 325
If file conflicts occur:
- Distinctive: line 326
Read current file state
- Distinctive: line 337
Read TASKS.md → Next ID is 015
- Distinctive: line 393
Read task015_add_password_reset.md
- Distinctive: line 423
When maintaining this template or adding new features, ALWAYS consult the .ai_platform_architecture/ folder for platform-specific …
- Very specific: line 444
All other platforms use .md files
- Distinctive: line 449
All other platforms: ❌ No equivalent (use rules/instructions instead)
- Distinctive: line 467
When adding features, ensure compatibility:
- Distinctive: line 470
Document compatibility: Update platform-specific files
- Very specific: line 472
Update comparison: Add to PLATFORM_COMPARISON.md table
- Distinctive: line 473
Migration guides: Update if file structure changes
- Distinctive: line 492
When maintaining this template:
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.
Files
fstrent Task Management Skill
Manage project tasks using the fstrent_spec_tasks task management system. This Skill provides capabilities for creating, updating, tracking, and organizing development tasks in a structured, file-based system compatible with multiple AI IDEs.
System Overview
The fstrent_spec_tasks system uses a file-based approach where:
- Master checklist:
.fstrent_spec_tasks/TASKS.md- Central task list with status indicators - Individual tasks:
.fstrent_spec_tasks/tasks/task{id}_descriptive_name.md- Detailed task files - Features:
.fstrent_spec_tasks/features/- Feature documentation referenced by tasks - Project context:
.fstrent_spec_tasks/PROJECT_CONTEXT.md- Project goals and scope
Task File Structure
File Naming Convention
- Format:
task{id}_descriptive_name.md - Examples:
task001_setup_database.md,task042_implement_auth.md - Location:
.fstrent_spec_tasks/tasks/
YAML Frontmatter Format
Every task file begins with YAML frontmatter containing metadata:
---
id: {number} # Sequential ID (001, 002, etc.)
title: 'Task Title' # Brief, actionable title
type: task|bug_fix|feature|retroactive_fix
status: pending|in-progress|completed|failed
priority: critical|high|medium|low
feature: Feature Name # Related feature (optional)
subsystems: [list] # Affected system components
project_context: 'Brief description of how this task relates to project goals'
dependencies: [task_ids] # Tasks that must complete first
estimated_effort: 'time estimate' # Optional: story points or hours
---Task Content Structure
After the YAML frontmatter, include:
- Objective: Clear, actionable goal
- Acceptance Criteria: Specific, measurable outcomes (checklist format)
- Implementation Notes: Technical details, approach, considerations
- Testing Plan: How to verify completion (optional)
- Resources Needed: Dependencies, documentation, tools (optional)
Task Operations
Creating a New Task
Process:
- Determine next available task ID by reading
.fstrent_spec_tasks/TASKS.md - Create task file:
.fstrent_spec_tasks/tasks/task{id}_descriptive_name.md - Add YAML frontmatter with all required fields
- Add task content (objective, acceptance criteria, notes)
- Update
.fstrent_spec_tasks/TASKS.mdwith new entry
TASKS.md Entry Format:
- [ ] Task {id}: {Brief description}Example:
- [ ] Task 001: Setup database schema
- [ ] Task 002: Implement user authenticationUpdating Task Status
Status Transitions:
pending→in-progress: Task work has startedin-progress→completed: Task successfully finishedin-progress→failed: Task blocked or abandoned- Any status →
pending: Reset if needed
Process:
- Update
statusfield in task file's YAML frontmatter - Update corresponding entry in
.fstrent_spec_tasks/TASKS.md
Status Indicators (Windows-safe emojis):
[ ]- pending (not started)[🔄]- in-progress (actively working)[✅]- completed (finished successfully)[❌]- failed (blocked or abandoned)
Example TASKS.md Updates:
- [ ] Task 001: Setup database schema
- [🔄] Task 002: Implement user authentication (in progress)
- [✅] Task 003: Create login page (completed)
- [❌] Task 004: Legacy system integration (blocked)Viewing Tasks
To view all tasks:
Read .fstrent_spec_tasks/TASKS.md for overview
To view specific task details:
Read .fstrent_spec_tasks/tasks/task{id}_*.md
To view tasks by status: Parse TASKS.md and filter by status indicators
To view project context:
Read .fstrent_spec_tasks/PROJECT_CONTEXT.md for goals and scope
Listing Tasks
Common queries:
- "Show all tasks" → Read TASKS.md
- "Show pending tasks" → Filter by
[ ] - "Show in-progress tasks" → Filter by
[🔄] - "Show completed tasks" → Filter by
[✅] - "Show task 5" → Read task005_*.md
Task Types
Standard Task (type: task)
Regular development work, features, improvements
Bug Fix (type: bug_fix)
Fixing defects or issues. Should reference BUGS.md entry.
Additional fields for bug fixes:
bug_reference: BUG-{number}
severity: critical|high|medium|low
reproduction_steps: 'How to reproduce'
expected_behavior: 'What should happen'
actual_behavior: 'What actually happens'Feature Task (type: feature)
Major feature implementation. Should reference feature document.
Link to feature:
feature: Feature Name # Must match .fstrent_spec_tasks/features/{feature-name}.mdRetroactive Fix (type: retroactive_fix)
Documents work completed in chat without prior task planning.
Additional fields:
created_date: '{completion_date}'
completed_date: '{completion_date}'
actual_effort: '{time_spent}'Task Complexity and Sub-Tasks
When to Create Sub-Tasks
If a task is complex (affects multiple subsystems, >2-3 days effort, many acceptance criteria), break it into sub-tasks.
Sub-Task Naming Convention
- Parent:
task042_implement_authentication.md - Sub-tasks:
task42.1_setup_auth_db.md,task42.2_create_auth_api.md,task42.3_add_auth_ui.md
Sub-Task YAML Format
---
id: "42.1" # String ID for sub-tasks
title: 'Setup Auth Database'
type: task
status: pending
priority: high
parent_task: 42 # Reference to parent
dependencies: [] # Can depend on sibling sub-tasks
---Task Dependencies
Specifying Dependencies
dependencies: [1, 3, 5] # Task IDs that must complete firstDependency Rules
- Tasks with dependencies should not start until dependencies are completed
- Check dependency status before starting work
- Update dependent tasks when blockers are resolved
Integration with Other Systems
Link to Features
Tasks reference features via the feature field:
feature: User Authentication # Links to features/user-authentication.mdLink to Bugs
Bug fix tasks reference BUGS.md:
bug_reference: BUG-001 # Links to entry in BUGS.mdLink to Project Context
All tasks should align with project goals in PROJECT_CONTEXT.md. The project_context field explains this connection.
File Organization Rules
Core Files (Always in .fstrent_spec_tasks/)
TASKS.md- Master task checklisttasks/- Individual task filesfeatures/- Feature documentationPLAN.md- Product requirementsPROJECT_CONTEXT.md- Project goalsBUGS.md- Bug tracking
Documentation (Goes in docs/)
- Project documentation
- API documentation
- Architecture diagrams
- Setup guides
- Migration files
Test Scripts (Goes in temp_scripts/)
- Test automation scripts
- Utility scripts
- Validation scripts
Auto-Creation Rules
When working with tasks, automatically create missing folders and files:
Auto-create folders:
.fstrent_spec_tasks/if it doesn't exist.fstrent_spec_tasks/tasks/if it doesn't existdocs/if needed for documentationtemp_scripts/if needed for test scripts
Auto-create template files:
TASKS.mdwith blank template if missingPROJECT_CONTEXT.mdwith template if missing
No confirmation needed - Create files and folders silently, report what was created.
Best Practices
Task Creation
- Use clear, actionable titles
- Include specific acceptance criteria
- Specify dependencies upfront
- Link to related features
- Estimate effort when possible
Task Updates
- Update status immediately when starting or completing
- Keep TASKS.md synchronized with task files
- Document blockers and issues
- Update acceptance criteria as work progresses
Task Organization
- Break complex tasks into sub-tasks
- Group related tasks by feature
- Track dependencies explicitly
- Archive completed tasks periodically (move to memory/)
Status Management
- Only one task should be
in-progressper developer at a time - Mark tasks
completedonly when all acceptance criteria are met - Use
failedstatus for blocked or abandoned work - Document reason for failure in task notes
Common Workflows
Workflow: Create Task from User Request
- User requests: "Create a task to implement user login"
- Read TASKS.md to determine next ID
- Create task file with proper naming
- Add YAML frontmatter with all fields
- Add objective and acceptance criteria
- Update TASKS.md with new entry
- Confirm task created with ID
Workflow: Update Task Status
- User requests: "Mark task 5 as in progress"
- Read task005_*.md file
- Update
status: in-progressin YAML - Update TASKS.md entry from
[ ]to[🔄] - Confirm status updated
Workflow: View Task Progress
- User requests: "Show me task progress"
- Read TASKS.md
- Count tasks by status
- Calculate completion percentage
- List in-progress tasks
- Identify blockers
Workflow: Complete Task
- User requests: "Complete task 3"
- Read task003_*.md
- Verify acceptance criteria are met
- Update
status: completedin YAML - Update TASKS.md entry from
[🔄]to[✅] - Optionally archive to memory/
- Confirm completion
Error Handling
Missing Files
If TASKS.md or task files are missing:
- Check if
.fstrent_spec_tasks/exists - Create directory structure if needed
- Create template files
- Inform user of initialization
Invalid Task IDs
If user references non-existent task:
- List available tasks
- Suggest correct ID
- Offer to create new task if intended
Concurrent Edits
If file conflicts occur:
- Read current file state
- Merge changes if possible
- Alert user to conflicts
- Suggest using version control
Examples
Example: Create Simple Task
User: "Create a task to add password reset feature"
Action:
- Read TASKS.md → Next ID is 015
- Create
.fstrent_spec_tasks/tasks/task015_add_password_reset.md:
---
id: 015
title: 'Add Password Reset Feature'
type: feature
status: pending
priority: high
feature: User Authentication
subsystems: [auth, email, ui]
project_context: 'Enables users to recover accounts, improving user experience and reducing support burden'
dependencies: [12]
estimated_effort: '1-2 days'
---
# Task 015: Add Password Reset Feature
## Objective
Implement password reset functionality allowing users to securely reset forgotten passwords via email.
## Acceptance Criteria
- [ ] User can request password reset from login page
- [ ] System sends reset link to user's email
- [ ] Reset link expires after 24 hours
- [ ] User can set new password via reset link
- [ ] Old password is invalidated after reset
- [ ] User receives confirmation email after reset
## Implementation Notes
- Use secure token generation for reset links
- Store tokens in database with expiration
- Send emails via existing email service
- Add UI for reset request and password change
- Update authentication system to handle resets
## Testing Plan
- Test reset request flow
- Verify email delivery
- Test token expiration
- Test password update
- Verify old password invalidation- Update TASKS.md:
- [ ] Task 015: Add password reset feature- Confirm: "Created Task 015: Add Password Reset Feature"
Example: Update Task Status
User: "Start working on task 15"
Action:
- Read task015_add_password_reset.md
- Update YAML:
status: in-progress - Update TASKS.md:
[🔄] Task 015: Add password reset feature - Confirm: "Task 015 status updated to in-progress"
Example: Complete Task
User: "Task 15 is done"
Action:
- Read task015_add_password_reset.md
- Review acceptance criteria
- Update YAML:
status: completed - Update TASKS.md:
[✅] Task 015: Add password reset feature - Confirm: "Task 015 marked as completed. Great work!"
Compatibility Notes
This Skill is designed to work with the same file format used by Cursor's fstrent_spec_tasks rules system. Tasks created in Claude Code can be viewed and updated in Cursor, and vice versa. The system uses:
- Standard YAML frontmatter
- Markdown content
- Windows-safe emoji indicators
- Git-friendly plain text files
Teams can use both IDEs interchangeably without workflow disruption.
Template Maintenance and Platform Architecture
Platform Architecture Reference
When maintaining this template or adding new features, ALWAYS consult the .ai_platform_architecture/ folder for platform-specific requirements and compatibility considerations.
Key Documentation:
- .ai_platform_architecture/PLATFORM_COMPARISON.md - Cross-platform comparison table
- .ai_platform_architecture/CLAUDE_CODE.md - Claude Code specific architecture
- .ai_platform_architecture/CURSOR.md - Cursor specific architecture
- .ai_platform_architecture/README.md - Overview and maintenance schedule
When to Reference Platform Architecture
ALWAYS check .ai_platform_architecture/ before:
- Adding new Skills, SubAgents, or Commands
- Modifying task file formats or naming conventions
- Creating new rules or instructions
- Updating file organization structure
- Making changes that affect cross-IDE compatibility
Critical Platform Differences
File Formats:
- Cursor uses
.mdcfiles (UNIQUE - won't work on other platforms) - All other platforms use
.mdfiles - Task files must use YAML frontmatter for cross-platform compatibility
Skills and SubAgents:
- Claude Code: ✅ Has Skills & SubAgents
- All other platforms: ❌ No equivalent (use rules/instructions instead)
Commands:
- Claude Code:
/command(slash prefix) - Cursor:
@command(@ prefix) - Others: See platform-specific docs
Periodic Verification
Quarterly Review (Every 3 Months):
- Check all platform official documentation for updates
- Test template on each platform
- Update
.ai_platform_architecture/documentation - Update PLATFORM_COMPARISON.md comparison table
- Document any breaking changes
Adding Features Cross-Platform
When adding features, ensure compatibility:
- Test on multiple platforms: Claude Code and Cursor minimum
- Document compatibility: Update platform-specific files
- Provide fallbacks: For platform-specific features
- Update comparison: Add to PLATFORM_COMPARISON.md table
- Migration guides: Update if file structure changes
Platform-Specific vs Universal Features
Universal (work on all platforms):
.fstrent_spec_tasks/task management system- Markdown documentation
- Basic file organization
- Standard project structure
- YAML frontmatter in task files
Platform-Specific (limited availability):
- Skills/SubAgents: Claude Code only
- .mdc files: Cursor only
- Command prefixes: Vary by platform (
/vs@) - MCP UI: Implementation varies by platform
Template Maintenance Workflow
When maintaining this template:
- Check platform docs in
.ai_platform_architecture/ - Identify compatibility requirements from PLATFORM_COMPARISON.md
- Test on primary platforms (Claude Code + Cursor)
- Update platform-specific folders (.claude/, .cursor/)
- Update documentation in
.ai_platform_architecture/ - Verify cross-platform compatibility
Resources
- Official Platform Docs: See
.ai_platform_architecture/README.mdfor links - Comparison Table:
.ai_platform_architecture/PLATFORM_COMPARISON.md - Migration Guides: In PLATFORM_COMPARISON.md
- Verification Status:
.ai_platform_architecture/README.mdstatus table
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
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.

