Menu
Use when crafting prompts for AI tools, managing context in AI conversations, or decomposing tasks for AI
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Execute safe Git workflows — branching, committing, resolving conflicts, and managing PRs
Use when controlling AI spend, token budgets, model routing, or workflow efficiency before scaling usage
Use when handling incidents, outages, severe regressions, or operational emergencies before attempting broad fixes
Use when investigating latency, throughput, resource saturation, or performance regressions before changing implementation details
Use when reviewing code, preparing a PR for review, or processing review feedback
Use when diagnosing bugs, test failures, or unexpected behavior before attempting any fix
| name | prompting |
| description | Use when crafting prompts for AI tools, managing context in AI conversations, or decomposing tasks for AI |
Announce at start: "Following the prompting skill for effective AI interaction."
Task first, context after. Lead with what you want, then provide supporting information.
[Specific, actionable task description]
Context:
- [Relevant code or file references]
- [Constraints and boundaries]
- [Expected behavior]
| Signal | Action |
|---|---|
| Task mentions "and" multiple times | Split on each "and" |
| Touches more than 2-3 files | Do one file at a time |
| Multiple behaviors to implement | One behavior per prompt |
| "While you're at it..." | Separate prompt |
[Task description]
Constraints:
- Only modify: [specific files or functions]
- Must preserve: [existing behavior, API, tests]
- Don't use: [forbidden dependencies, patterns]
- Must support: [versions, environments, edge cases]
| Do | Don't |
|---|---|
| Share the specific functions mentioned in the error | Dump the entire codebase |
| Include relevant types and interfaces | Include unrelated modules |
| Show error messages and stack traces | Describe errors vaguely |
| Reference existing patterns to match | Assume AI knows your conventions |
| Symptom | Cause | Fix |
|---|---|---|
| AI forgets instructions | Context pushed out | Repeat key constraints |
| AI contradicts previous output | Lost history | Start fresh session |
| AI ignores shared code | Code too far back | Re-share relevant snippets |
| Generic responses | Overloaded context | Reduce scope |
Fix this error: [full error message]
Stack trace: [full stack trace]
Code: [relevant function]
Expected: [correct behavior]
Actual: [current behavior]
Add [feature] to [component].
Behavior:
- When [trigger], it should [action]
Constraints:
- [Scope limits]
- [What must not change]
Match the pattern in [existing similar code].
| When | Invoke |
|---|---|
| Prompting for a multi-step task | planning |
| Prompting for debugging help | debugging |
| Prompting for code review | code-review |
For principles, rationale, anti-patterns, and examples:
guides/prompting-patterns/prompting-patterns.mdguides/context-management/context-management.md