{"id":"ckelsoe-prompt-architect","name":"prompt-architect","homepage":"https://github.com/ckelsoe/prompt-architect","repo_url":"https://github.com/ckelsoe/prompt-architect","category":"devtools","subcategories":[],"tags":["devtools","agent-skills","prompt-engineering","prompt-optimization","cli","prompt-frameworks","llm-coding-agents"],"what_it_does":"CLI/Agent Skills-compatible tool that transforms vague prompts into structured, expert-level prompts using a set of research-backed prompt frameworks, optionally asking clarifying questions and iteratively refining with before/after comparisons.","use_cases":["Converting vague task requests into high-quality prompts for LLM coding assistants (Claude Code, Cursor, Copilot, etc.)","Prompt engineering for content writing, planning, data transformations, and constraint-driven tasks","Generating reusable prompt templates/frameworks tailored to an intent category","Improving existing prompts or rewriting prompts using a structured before/after/bridge approach"],"not_for":["Production systems that need a stable external HTTP API","Use cases requiring guaranteed factual correctness from external sources (this is prompt structuring, not retrieval)","Use cases that require formal compliance guarantees without human review"],"best_when":"You want to consistently structure prompts for agentic coding/content workflows and reduce ambiguity via guided questions and framework selection.","avoid_when":"You need an automated, authenticated service endpoint with well-defined API contracts, rate limits, and idempotent operations.","alternatives":["Manual prompt framework templates (CO-STAR/RISEN/etc.)","Other Agent Skills compatible “prompt optimizer” skills","Custom scripts/notebooks that apply prompt rubric-based transformations"],"af_score":33.5,"security_score":51.8,"reliability_score":35.0,"package_type":"skill","discovery_source":["openclaw"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T13:54:11.792216+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":false,"mcp_server_url":null,"has_sdk":false,"sdk_languages":["JavaScript","TypeScript","Python"],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":["GitHub token for npm/pkg registry access (read:packages)"],"oauth":false,"scopes":true,"notes":"Authentication mentioned is for installing from GitHub Packages (via .npmrc and GitHub token with read:packages), not for an API service."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"No pricing information provided; as an npm package, cost is likely via LLM usage in downstream agents, not the tool itself."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":33.5,"security_score":51.8,"reliability_score":35.0,"mcp_server_quality":0.0,"documentation_accuracy":60.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":75.0,"rate_limit_clarity":0.0,"tls_enforcement":70.0,"auth_strength":55.0,"scope_granularity":35.0,"dependency_hygiene":60.0,"secret_handling":40.0,"security_notes":"Security posture is partially inferable: publishing license is MIT and only one dependency is listed in the manifest (@clack/prompts). However, the provided content does not describe runtime network security, TLS enforcement, logging behavior, or how secrets (GitHub token) are handled during install. The authentication requirement is limited to package registry access rather than user data access. Validate that the tool does not log tokens and that it follows secure npm install practices.","uptime_documented":0.0,"version_stability":55.0,"breaking_changes_history":45.0,"error_recovery":40.0,"idempotency_support":"false","idempotency_notes":null,"pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["This appears to be a CLI/Agent Skills prompt transformer rather than a network API; typical agent expectations like rate limiting, pagination, and idempotency for HTTP calls may not apply.","The README emphasizes compatibility with multiple agent tools; behavior and input/output formats may differ across host tools, so an agent should validate the skill integration for the specific host environment."]}}