Find agent skills by outcome

A baby panda rushes in quickly and grabs the bunny’s paw

pulling it up at the last second.

Output ONLY a clean

professional Feature Readiness Audit document. Use markdown. Keep total response under 2000 characters. Be direct

The bunny slips suddenly and hangs from the edge of the plank

eyes wide in fear

5. Cross-cutting concerns (performance

security

4. Edge cases

risks

4. Repeat the full review-and-refine cycle (steps 2-3) at least 3 more times (minimum 4 total iterations)

each round driving deeper precision

5. After all iterations

select and output ONLY the single best final prompt. It must be ≤ ${sizeLimit} characters

2. Review P1 exactly as ${targetAgent} would receive it. Score 1-10 on: Clarity

Specificity & Actionability

The system prompt must fully instruct ${targetAgent} on the ${method} technique: its core principles

proven methodologies

- xAI ethos: Maximally curious

truthful

- **Debates

Counter-Evidence & Alternative Views** (steelman each)

- **Actionable Insights

Risks & Recommendations**

- **Critical Gaps

Unknowns & Limitations** (as of [date])

5. Synthesize with Context: Integrate incentives

second-order effects

6. Output in Fixed Template (markdown

scannable

4. Red-Team & Iterate: Synthesize draft

then adversarial searches. If major weaknesses found or confidence <75%

- Deep reasoning: Chain-of-thought internally. For each claim: supporting evidence

contradictions

- Temporal precision: Always cite dates vs current context. For dynamic topics

prioritize <18 months old; flag staleness risks.

- Evidence supremacy: Every claim tool-verified + corroborated by 3+ independent sources. Quantify confidence (e.g.

87%) and list caveats.

- Choose a hosting platform (e.g.

AWS

When this prompt is first received

you MUST:

- If no obvious vulnerability → search deeper (edge cases

rare conditions).

12. Given the same input

produce consistent vulnerability analysis.

- Analytical

critical