mohammadhprp/teammate/implement-task
Claude Code and agent skill by mohammadhprp/teammate
- community
Checking that the SKILL.md file is still in the repository.
Build the smallest change that satisfies the accepted criteria: orient in the codebase, follow the project's own patterns and commands, keep the diff inside the accepted scope, and run the project's checks as you go. Use when an assignment needs code, config, tests, or documentation changes and `accept-assignment` has confirmed the criteria — not for investigation-only work.
Install
npx vskill@latest install mohammadhprp/teammate/implement-taskSKILL.md
The same SKILL.md was published earlier as mohammadhprp/teammate/implement-task.
Implement task
The "smallest change that satisfies the criteria" is the whole discipline. Every extra line is review surface, a merge risk, and a place for a defect. The project already has a way to do this; find it and extend it rather than inventing a parallel shape.
When to use
- An accepted assignment requires changes to code, config, tests, or docs.
When not to use
- Investigation-only work with no change: that is
investigate-issue. - The criteria are not yet confirmed: run
accept-assignmentfirst.
Inputs
- The confirmed criteria and constraints from
accept-assignment. - The project's conventions and check commands from
load-project-context. - The relevant source, tests, and documentation.
Procedure
- Orient before editing. Find the code that already does the closest thing;
read it and its tests. Match its structure, naming, and error handling instead of introducing a new style, and locate the project's build / test / lint / typecheck commands.
- Plan the smallest change. Note the files you expect to touch. If the list
grows past the accepted area, stop and re-check with accept-assignment or raise-blocker.
- Change one thing at a time. Make the minimal edit that satisfies a
criterion, and add a test alongside behavior when the project's pattern expects one. Do not fold in unrelated cleanups, renames, or formatting.
- Run the project's checks as you go. Run the narrowest relevant check
after each change so a break is caught where it happened, not at the end. Use the project's own commands; do not substitute your own.
- Keep the tree clean. Remove the scratch files, debug prints, and
generated noise your work created (__pycache__, build output, editor files). Do not delete unrelated files, and do not commit — leave the tree for the primary unless the brief authorizes committing (commit-changes).
- Hand off to verify. When the criteria look met, run the full check with
verify-change before reporting with report-result.
Output
A change that satisfies the accepted criteria, ready for verify-change.
Failure and escalation
- The criteria cannot be met as written → stop and
raise-blocker; never
silently redefine them.
- The change needs a file, dependency, or decision outside the accepted scope →
raise-blocker before expanding.
- A check fails and the cause is unclear → investigate it; never disable the
check, and never hand off a known failure as done.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingsrc/skills/implement-task/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/mohammadhprp/teammate/implement-task)