implement_direct
Skill21 starsChanged 3 months agoMITView on GitHub
Implement a small issue directly without planning phase
In this repository21 other files it gives its agents.
- .claude/CLAUDE.mdCLAUDE.md
- check_branch_statusSkill
- commit_pushSkill
- discussSkill
- implementation_approveSkill
- implementation_finaliseSkill
- implementation_needs_reworkSkill
- implementation_new_tasksSkill
- implementation_reviewSkill
- implementation_review_supervisorSkill
- issue_analyseSkill
- issue_analyse_supervisorSkill
- issue_approveSkill
- issue_createSkill
- issue_requirementsSkill
- issue_updateSkill
- plan_approveSkill
- plan_reviewSkill
- plan_review_supervisorSkill
- plan_updateSkill
- rebaseSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 6 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 23
If no issue number is provided:
- Very specific: line 24
Read .vscodeclaude_status.txt and extract the issue number from the Issue #NNN line
- Distinctive: line 25
If the file doesn't exist or has no issue number, ask the user
- Very specific: line 30
Call mcp__mcp-workspace__github_issue_view with the issue number.
- Distinctive: line 31
Read the issue title, description, and acceptance criteria carefully.
- Distinctive: line 39
Read relevant source files referenced in the issue
- Distinctive: line 45
Follow existing code patterns and conventions
- Distinctive: line 46
Keep changes focused and minimal — only what the issue requires
- Very specific: line 55
Use mcp__mcp-tools-py__run_format_code to format all code (black + isort).
≈ “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
Implement Direct
Implement a small, well-defined issue directly — no planning phase, no task tracker.
Resolve Issue Number
The user may provide an issue number as the argument (available as $ARGUMENTS).
If no issue number is provided:
- Read
.vscodeclaude_status.txtand extract the issue number from theIssue #NNNline - If the file doesn't exist or has no issue number, ask the user
Steps
- Fetch issue details
Call
mcp__mcp-workspace__github_issue_viewwith the issue number. Read the issue title, description, and acceptance criteria carefully. - Checkout/create issue branch
mcp-coder gh-tool checkout-issue-branch <issue_number> - Understand context
- Read relevant source files referenced in the issue
- Understand the existing code patterns and conventions
- Identify the minimal set of changes needed
- Implement changes
- Make the required code changes directly
- Follow existing code patterns and conventions
- Keep changes focused and minimal — only what the issue requires
- Run quality checks
mcp__mcp-tools-py__run_pylint_check— fix all issuesmcp__mcp-tools-py__run_pytest_check(withextra_args: ["-n", "auto"]) — fix all failuresmcp__mcp-tools-py__run_mypy_check— fix all issuesmcp__mcp-tools-py__run_ruff_check— fix all issues
- Format code
Use
mcp__mcp-tools-py__run_format_codeto format all code (black + isort). - Update issue status
mcp-coder gh-tool set-status status-07:code-review - Suggest follow-up steps
/commit_push— commit and push changes/check_branch_status— verify branch is clean/implementation_review— request a review of the implementation
Scope Guidance
This skill is designed for small, well-defined issues that can be implemented in a single pass. If the issue is complex or involves multiple components, recommend the full workflow instead:
/create_plan— create a detailed implementation plan/implement— implement the plan step by step
Note: This skill has disable-model-invocation — it can only be run by the user typing /implement_direct.
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.

