spec-develop
Skill16 starsChanged 4 days agoMITView on GitHub
Execute an OpenSpec tasks.md during implementation. TRIGGER when: working through spec task checkboxes and applying code changes. SKIP: authoring the tasks.md (use spec-design); verifying the result (use spec-review).
In this repository39 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- agent-commsSkill
- agent-memorySkill
- api-designSkill
- context-engineeringSkill
- database-optimizationSkill
- distillSkill
- docker-templatesSkill
- error-handlingSkill
- frontend-designSkill
- github-actions-templateSkill
- git-operationsSkill
- logging-standardsSkill
- mcp-toolsSkill
- monitoring-observabilitySkill
- mui-stylingSkill
- necessity-firstSkill
- pattern-recognitionSkill
- planning-methodologySkill
- python-patternsSkill
- quality-validationSkill
- react-patternsSkill
- research-methodologySkill
- reverse-engineer-anythingSkill
- security-review-checklistsSkill
- semantic-memory-mcpSkill
- skill-authoringSkill
- sofa-searchSkill
- spec-designSkill
- spec-researchSkill
- spec-reviewSkill
- spec-workflowSkill
- state-managementSkill
- systematic-debuggingSkill
- testing-strategySkill
- ui-ux-pro-maxSkill
- verification-before-completionSkill
- watch-patternsSkill
- worktree-managementSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
Read the task description and target file path
- Distinctive: line 33
Check design.md for relevant requirements and decisions
- Distinctive: line 47
If a task cannot be completed:
- Very specific: line 84
If tasks.md has some [x] items already:
- Distinctive: line 91
When design.md specifies a technical choice, follow it:
≈ “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.
Files
OpenSpec Apply (Task Execution)
Guide for implementing code changes driven by tasks.md.
Input Files
| File | Purpose | Action |
{specs_path}/tasks.md | Implementation checklist | Read, execute, mark done |
{specs_path}/design.md | Architecture decisions | Reference during implementation |
{specs_path}/proposal.md | Context and motivation | Reference if unclear on intent |
Path Enforcement: The specs_path MUST be .scaffolding/conversations/{UUID}/specs/ where {UUID} is a valid UUID (format: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx). NEVER use descriptive folder names.
.scaffolding/conversations/ is a symlink to <share>/projects/<repo-name>/conversations (paths below are unchanged; always go through the symlink).
Execution Protocol
Step 1: Read tasks.md
Parse all task groups and checkboxes. Identify dependency order.
Step 2: Execute tasks in order
For each task (- [ ] N.M description):
- Read the task description and target file path
- Check design.md for relevant requirements and decisions
- Implement the change following project patterns
- Mark complete: change
- [ ]to- [x]
Step 3: Run validation after each group
| Stack | Validation Command |
| Backend | pytest |
| Frontend | npm run validate |
| Both | Run both commands |
Step 4: Handle blockers
If a task cannot be completed:
- [ ] 2.3 Implement caching layer
> BLOCKED: Redis connection config not yet available. Depends on devops setup.Continue with unblocked tasks. Do not stop the entire chain.
Rules
| Rule | Detail |
| Execute in order | Respect dependency sequence |
| Mark progress | - [ ] -> - [x] as you complete |
| Reference design | Check design.md before implementing |
| Never modify specs | Only modify tasks.md checkboxes, never change spec content |
| Validate per group | Run tests after each task group |
| Note blockers | Add > BLOCKED: note, skip and continue |
| Search before write | Check core/utils/, core/exceptions.py before creating helpers |
Progress Tracking
Checkpoint format in tasks.md
## 2. Core Implementation
- [x] 2.1 Implement service in `./backend/service.py`
- [x] 2.2 Add Pydantic schemas in `./backend/schemas.py`
- [ ] 2.3 Add router in `./backend/router.py`
> BLOCKED: Waiting for auth middleware decision
- [x] 2.4 Run validation: `pytest`Resuming after interruption
If tasks.md has some [x] items already:
- Skip completed tasks
- Resume from first unchecked
- [ ] - Re-read design.md for context on remaining work
Integration with Design Decisions
When design.md specifies a technical choice, follow it:
| Design says | You do |
| "Use Repository pattern" | Create repository class, not inline queries |
| "Use Pydantic v2 model_validator" | Use @model_validator, not @validator |
| "Max 500 lines per file" | Split if approaching limit |
| "Existing utility in core/utils" | Import it, do not recreate |
Anti-Patterns
| Avoid | Instead |
| Skipping validation steps | Run tests after every group |
| Modifying design.md content | Only reviewer or planner changes specs |
| Implementing out of order | Follow dependency sequence |
| Ignoring blocked tasks silently | Add explicit > BLOCKED: note |
| Creating duplicate utilities | Search core/ first |
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.