| name | dummy-test-skill |
| description | Use when [TRIGGER: specific symptoms or situations]... |
Dummy Test Skill
Purpose: [Detailed explanation of WHY this skill exists and WHAT problem it solves]
1. Red Flags (Anti-Rationalization)
STOP and READ THIS if you are thinking:
- "[Common excuse 1]" -> [Why this is wrong — explain the consequence]
- "[Common excuse 2]" -> [Why this is wrong — explain the consequence]
2. Capabilities
List what this skill allows the agent to do.
- [Capability 1]
- [Capability 2]
3. Execution Mode
- Mode: [prompt-first | script-first | hybrid]
- Why this mode: [Short rationale for this skill]
4. Script Contract
Use this section when mode is script-first or hybrid.
- Command(s):
python3 scripts/[tool].py --arg value
- Inputs: [required args/files/config]
- Outputs: [artifacts/logs/exit behavior]
- Failure semantics: [non-zero exit, error schema]
- Idempotency: [how repeated runs behave]
- Dry-run support: [yes/no and flag]
5. Safety Boundaries
- Allowed scope: [path/module/target constraints]
- Default exclusions: [system/non-target areas]
- Destructive actions: [must be explicit opt-in, never default]
- Optional artifacts: [whether absence is blocking or non-blocking]
6. Validation Evidence
- Local verification: [commands/checks]
- Expected evidence: [files, logs, structured output]
- CI signal: [job/check name if available]
7. Instructions
Provide step-by-step instructions. Use graduated language:
- Safety-critical steps:
MUST + explain why
- Behavioral steps: Explain why + imperative verb
[Phase/Action 1]
- Step 1: [Instruction]
- Step 2: [Instruction]
- Tip: [Sub-instruction or nuance]
[Phase/Action 2]
...
8. Workflows (Optional)
For complex tasks, use a checklist:
- [ ] Analysis
- [ ] Execution
- [ ] Verification
9. Best Practices & Anti-Patterns
| DO THIS | DO NOT DO THIS |
|---|
| [Best practice] | [Common mistake] |
| [Best practice] | [Common mistake] |
Rationalization Table
| Agent Excuse | Reality / Counter-Argument |
|---|
| "[Excuse 1]" | "[Counter-argument]" |
| "[Excuse 2]" | "[Counter-argument]" |
10. Examples (Few-Shot)
Refer to examples/ directory for full files, or include short snippets here.
Input:
[User Request]
Output:
[Ideal Agent Response/Action]
11. Resources
Describe the usage of files in assets/, references/, and scripts in scripts/.
scripts/helpers.py: [Description]
assets/template.json: [Description]
references/guidelines.md: [Description]
12. Evals (Optional but Recommended)
Define 2-3 test prompts in evals/evals.json — see skill-creator §14 for format.