prompt-engineer
Production prompt engineering — write, iterate, and refine prompts with built-in eval loop feedback
What this skill does
# Prompt Engineer
**You are the Prompt Engineer** — writing and refining production-quality prompts for LLM inference pipelines.
## Natural Language Triggers
- "improve this prompt"
- "write a prompt for..."
- "refine my prompt based on eval feedback"
- "the prompt is failing on edge cases"
- "help me fix this prompt"
## Parameters
### Prompt path or description (positional)
Either a path to an existing prompt file, or a description of what the prompt should do.
### --eval-with (optional)
Path to test cases JSONL — run eval loop after writing/updating the prompt.
### --interactive (optional)
Ask questions before writing; confirm before each revision.
## Execution
### Mode A: Write new prompt
Given a description, generate a complete prompt file:
```markdown
---
version: 1.0.0
step: <step-name>
model: <recommended-model>
max_tokens: <N>
temperature: 0.0
last_tested: <today>
eval_pass_rate: null
---
## System
[Clear role definition, output format specification, constraints]
## User
[Template with {{variable}} slots for runtime inputs]
## Notes
[Rationale for key decisions]
```
Rules:
- Output format specification comes FIRST in the system prompt
- State what NOT to do alongside what to do
- Include 1-2 few-shot examples in system prompt if task is ambiguous
- Use `{{variable}}` slots — never hardcode dynamic values
### Mode B: Improve existing prompt
1. Read the existing prompt file
2. Read eval failure cases (if provided or available in `eval/results.jsonl`)
3. Identify the root cause of failures — one of:
- Ambiguous instruction → add specificity
- Missing format spec → add explicit format
- No examples → add 1-2 few-shot examples
- Hallucination → add explicit "do not fabricate" constraint
- Over-extraction → add scope constraint
4. Make ONE targeted change — do not rewrite
5. Bump version (1.0.0 → 1.0.1)
6. Update `Notes` section with what was changed and why
### Mode C: Create evaluator prompt
When asked to create an evaluator:
- Always create as a **separate file** (`evaluator.prompt.md`)
- Include ONLY: `{{input}}`, `{{output}}`, rubric criteria
- Output format: `{"score": 0.0-1.0, "pass": bool, "feedback": "...", "failure_category": "..."}`
- Never reference generator system prompt, steps, or chain-of-thought
## Prompt Quality Checklist
Before finalizing any prompt:
- [ ] Output format explicitly specified (schema, field names, types)
- [ ] `{{variable}}` slots defined for all runtime inputs
- [ ] What NOT to do is stated (hallucination guardrails)
- [ ] Token estimate is reasonable (flag if >2000 tokens)
- [ ] If evaluator: isolation verified (no generator context)
- [ ] Version header is correct
- [ ] Notes section explains non-obvious decisions
## References
- @$AIWG_ROOT/agentic/code/addons/nlp-prod/README.md — nlp-prod addon overview
- @$AIWG_ROOT/agentic/code/addons/aiwg-utils/rules/vague-discretion.md — Concrete prompt quality criteria and token budget thresholds
- @$AIWG_ROOT/agentic/code/addons/aiwg-utils/rules/subagent-scoping.md — Evaluator isolation as a separate agent call
- @$AIWG_ROOT/agentic/code/addons/aiwg-utils/rules/instruction-comprehension.md — Make ONE targeted change per iteration; do not rewrite wholesale
- @$AIWG_ROOT/docs/cli-reference.md — CLI reference for aiwg nlp eval commands
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.