Browse 20,400 Prompts

Auto-generates high-quality prompts from a brief description

Side-by-Side (SxS) interleaved reasoning strategist — designs when an agent should reveal reasoning vs. keep it private in streaming interfaces; support-threshold gating, update-granularity ladders, silence-tax management, anti-filler rules, correction protocols for commitment bias; based on "When to Think, When to Speak" (arXiv 2605.03314, ICML 2026)

Data-management-first memory system architect — designs representation/storage, extraction, retrieval/routing, and maintenance as measurable modules; workload-aware benchmarking, localized-vs-global maintenance trade-offs, update-correctness discipline; based on "Are We Ready For An Agent-Native Memory System?" (arXiv 2606.24775, June 2026; OpenDataBox/MemoryData benchmark suite)

Post-deployment trajectory sampling and triage prompt — three-dimensional signal taxonomy (interaction / execution / environment), cheap-rules-first extractors, diversified ranking, reviewer-feedback loop, explicit privacy-redaction step; designed to lift informative traces over random sampling without ground-truth labels; based on "Signals: Trajectory Sampling and Triage for Agentic Interactions" (arXiv 2604.00356, April 2026, 6.2k HF likes)

Cost-efficient routing strategist for LLM-as-a-Judge — per-query decisions between reasoning and non-reasoning judges under a hard budget, task-class decomposition (VERIFICATION / PREFERENCE / AMBIGUOUS), leakage-safe routing signals, KL-ball distributionally-robust optimization, budget accounting with end-of-window carve-out, production drift monitoring with rho-widening, "reasoning theater" detection on simple items, mandatory pre-promotion Pareto-dominance check against always-reaso

Agent context database architect — filesystem-paradigm unification of memories, resources, and skills; L0/L1/L2 tiered loading, directory recursive retrieval, visualized trajectories, and session-based memory iteration; based on volcengine/OpenViking (Jan 2026, 26.8k+ stars, AGPLv3)

Unified four-layer architect that decides which cognition stays in weights, which lives in the prompt, and which is externalized into memory / skills / protocols / harness — precondition check, per-layer audit (what belongs where, what does not), interface contracts between layers (no cross-layer bypass), invariants (separation of concerns / least privilege / inspectability / reversibility / versioning), test plan, and a strict output contract that forces every cognitive function to de

Governed knowledge-vault architect beneath RAG — typed Markdown artifacts, deterministic set-algebraic selectors, contextnest:// URI citations, SHA-256 hash-chained versions, graph checkpoints, MCP source nodes, and audit traces so every agent output is reconstructible; based on ContextNest (arXiv 2607.02116, July 2026; IBM Research / Emory)

Curated Coding prompt from awesome-prompts library split.

Prompt for multi-step agents that must absorb mid-task user changes safely — state snapshot, stop/preserve decisions, re-plan, irreversible-risk tracking (2026)

Treats JSON Schema / Pydantic / function-calling schemas as a second instruction channel — audits instruction-silent keys ("output", "result", "data"), reorders scaffolding-before-conclusion, rewrites descriptions as inline directives, lifts prose constraints into enums/shapes/cardinality, versions schema diffs as prompt diffs, and probes fragility with no-change-expected vs change-expected edits; based on "Schema Key Wording as an Instruction Channel in Structured Generation" (arXiv 2

Prompt for designing, evaluating, and iteratively improving agent tools — tool selection/omission (constraint collapse), namespacing, context-rich returns, token-efficient responses, description prompt-engineering, agent-driven optimization loops; based on Anthropic's 2026 "Writing effective tools for agents" guidance

Prompt for retrieval/synthesis/critique coordination — evidence tables, stop conditions, conflict handling, confidence tracking in multi-agent RAG workflows (2026)

Production prompt extracted from developer Python & TypeScript codebases.

Three-layer failure-mode auditor for web/GUI/computer-use agents — separates planning, grounding, and replanning failures with quoted-evidence localisation; default grounding-blame prior (per the paper, grounding dominates), one-exploratory-replan-per-failure rule, PDDL-vs-NL plan validation, upstream rule-out (auth, captcha, prompt injection, goal underspec), layer-targeted fix bucketing, mandatory pre/post-fix regression probe; based on Why Do Web Agents Fail? A Hierarchical Planning

Prompt for reviewing agent systems across control, ambiguity handling, security, transparency, and privacy — based on Anthropic's 2026 trustworthy-agent guidance

Prompt for designing interactive UI extensions for MCP servers — `ui://` resources, `_meta.ui` tool bindings, sandboxed iframe bridge, JSON-RPC over postMessage, permissions/CSP; based on the MCP Apps open standard (Anthropic/OpenAI, 2026)

Provider-neutral agent harness architect — MVP blueprint, loop design, tool/permission contracts, context/memory/compaction, planning/goals, skills/MCP connectors, prompt caching, observability/evals, safety guardrails; based on DenisSergeevitch/agents-best-practices (May 2026, 654 stars)

Production prompt extracted from developer Python & TypeScript codebases.

Agent-designing-agent prompt for creating reusable, self-evaluating skills — Read-Execute-Reflect-Write loop, SKILL.md scaffolding, versioned skill libraries (2026)

Runtime substrate that turns agent execution into a first-class, inspectable object — typed events for model/tool/environment changes, Git-like trace with deterministic fork/replay/intervene primitives, 5× faster fork than Docker commit; based on Stanford's "Shepherd: A Runtime Substrate Empowering Meta-Agents with a Formalized Execution Trace" (arXiv 2605.10913, May 2026)

Prompt engineering for non-autoregressive diffusion language models (LLaDA, Dream, MMaDA) — bidirectional prefix/suffix conditioning, fill-in-the-middle design, mask scheduling, step-level intervention, test-time scaling via S³ parallel trajectories + verifier selection, CFG and temperature analog tuning; based on 2025–2026 diffusion-LM research (2026)

Stop-and-think operator for complex tool-use chains — dedicated `think` tool checkpoints to interpret tool outputs, verify policy compliance, and decide next actions; based on Anthropic's "The think tool: Enabling Claude to stop and think" (Aug 2026)

Zero-waste prompt engineer for any AI tool — 9-dimension intent extraction, 20+ tool-specific profiles (Claude 4.x, GPT-5.x, o3, Gemini 3, Cursor, Midjourney, ComfyUI), diagnostic checklist, token-efficiency audit; based on nidhinjs/prompt-master (Mar 2026)