Prompt and context engineering resources

Rules, guides, commands, and skills for prompt design, CLAUDE.md setup, context hygiene, instructions, and reusable agent memory.

Rules, guides, commands, and skills for prompt design, CLAUDE.md setup, context hygiene, instructions, and reusable agent memory.

Compared at a glance

The top 5 picks side by side on trust, install, platform support, and disclosed notes — full rationale for each below.

3 trust signals differ across this comparison (Package trust, Source provenance, Submitter).

Next steps differ across picks — use the actions in the table below to copy install commands and source links per resource.

Field	Bootstrap CLAUDE.md with /init in Claude Code Use the built-in /init command to generate a project CLAUDE.md from your codebase, then refine it by hand. Covers /memory, file locations, @path imports, and .claude/rules/. Open dossier	Windsurf AI-Native Collaborative Development Skill Master collaborative AI-assisted development with Windsurf IDE's Cascade AI, multi-file context awareness, and Flow patterns for team workflows. Open dossier	Compaction and Memory Hygiene for Long Claude Sessions Guide to /compact, /memory, and CLAUDE.md hygiene for long Claude Code sessions: when to compact, what to store in memory, and avoiding stale context. Open dossier	Manage Prompt and Context Hygiene in Long Coding Sessions A practical guide for keeping Claude Code sessions focused during long coding work with scoped prompts, checkpoints, durable memory boundaries, source refreshes, and privacy-safe handoffs. Open dossier	Codex Automations Orchestrator Capability Pack Skill Expert automation-orchestration capability pack for designing safe, low-noise recurring Codex workflows with clear runbooks. Open dossier
Next stepsDiffers	Open dossierAPI JSONOpen LLMOpen sourceNewsletterClaim listing	Open dossierCopy installAPI JSONOpen LLMOpen sourceNewsletterClaim listing	Open dossierAPI JSONOpen LLMOpen sourceNewsletterClaim listing	Open dossierAPI JSONOpen LLMOpen sourceNewsletterClaim listing	Open dossierCopy installAPI JSONOpen LLMOpen sourceNewsletterClaim listing
Trust	Review first	Trusted	Review first	Review first	Trusted
Review status	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed	ReviewedMaintainer reviewed
Package trustDiffers	Package not verified	Package verified2025-10-16	Package not verified	Package not verified	Package verified2026-04-11
Source provenanceDiffers	Source-backed	No submission link	Submission linkedSource submission	Source-backed	No submission link
SubmitterDiffers	—	—	kiannidev	MkDev11	—
Install risk	Review first	Low risk	Review first	Review first	Low risk
Notes	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓	Safety ✓ Privacy ✓
Brand	—	Windsurf	—	—	—
Category	commands	skills	guides	guides	skills
Source	Source-backed	First-party	Source-backed	Source-backed	First-party
Author	JSONbored	JSONbored	kiannidev	MkDev11	JSONbored
Added	2025-10-25	2025-10-16	2026-06-14	2026-06-04	2026-04-11
Platforms	Claude Code	Claude CodeCodexWindsurfGeminiCursorCLI	Claude Code	Claude Code	Claude CodeCodexWindsurfGeminiCursorCLI
Harness	Claude Code	Claude CodeCodexWindsurfGeminiCursorCLI	Claude Code	Claude Code	Claude CodeCodexWindsurfGeminiCursorCLI
Source repo	—	—	—	—	—
Safety notes	✓/init writes a CLAUDE.md file to your project (and, in the CLAUDE_CODE_NEW_INIT=1 flow, may also propose skills and hooks). It analyzes your codebase before writing; review the generated file before committing. CLAUDE.md, CLAUDE.local.md, and .claude/rules/ are loaded into Claude's context every session as behavioral guidance, not enforced configuration. For hard enforcement (blocking tools/commands), use a PreToolUse hook or managed settings instead. Custom commands in .claude/commands/*.md are prompt templates committed to the repo and run with your session's permissions; review any command file before invoking it.	✓Setup downloads and unzips a package and changes Windsurf IDE configuration; review what it writes and run it in a trusted workspace.	✓Compaction summarizes conversation history; verify summaries before relying on them for security-sensitive or release-critical decisions. Do not store secrets, credentials, or customer identifiers in `/memory` or CLAUDE.md. After compaction, re-validate repository state, test results, and open PR status before continuing destructive edits.	✓Do not paste credentials, production secrets, private customer records, or unrelated proprietary files into prompts just to "fill context." Treat long-session conclusions as provisional when the repository, dependencies, docs, or issue state may have changed. Keep implementation authority separate from analysis: ask for review before applying broad file edits, running risky commands, or changing public project state.	✓Designs and runs automation workflows that can execute commands and trigger external actions; review each automation's permissions and triggers before enabling it.
Privacy notes	✓CLAUDE.md and .claude/rules/ are committed to source control and shared with your team, so do not put secrets, credentials, or sensitive data in them. Use CLAUDE.local.md (gitignored) for personal, machine-local notes. /init reads files across your project (and existing AGENTS.md/.cursorrules/.windsurfrules) to generate context; do not let it copy secret values into the generated CLAUDE.md. Auto memory is stored locally under ~/.claude/projects/<project>/memory/ and is machine-local; review it with /memory if you are concerned about what was recorded.	✓Windsurf (Cascade) sends your code and prompts to its AI backend to power multi-file operations; review Windsurf's data-handling settings and keep API keys and secrets out of committed config.	✓Memory files persist across sessions and may be committed if stored in project scope. Compaction summaries can retain file paths, issue titles, and internal URLs—review before sharing session exports. Clear memory when offboarding contractors or rotating shared machines.	✓Long coding sessions can accumulate source code, file paths, stack traces, logs, issue details, usernames, and internal decisions. Durable memory and project instructions can persist sensitive details longer than an ordinary prompt, so store only stable, non-secret facts there. Handoff summaries should mention decisions and next steps without copying full private logs, tokens, customer reports, or unnecessary code excerpts.	✓Inputs can include source files, prompts, logs, account metadata, repository details, and operational context that may be sent to the configured AI model. Redact secrets, customer data, private URLs, credentials, and proprietary implementation details before sharing prompts, reports, or generated artifacts.
Prerequisites	— none listed	Windsurf IDE installed (download from codeium.com/windsurf) Active project/codebase Git repository (recommended for version control) Codeium account (for Cascade AI features)	An active long-running Claude Code session or multi-day task spanning several sessions. A project with CLAUDE.md or memory files defining durable conventions. Agreement on what may be written to user or project memory versus kept ephemeral. Permission to run `/compact` and inspect resulting summaries before continuing work.	A Claude Code project or repository with enough scope that work may span several prompts or sessions. Agreement on what belongs in durable memory, project instructions, slash commands, or temporary conversation context. A habit of validating source URLs, test results, and repository state before relying on old assumptions. A place to record handoff summaries, decisions, and unresolved risks without storing secrets.	Automation goals and success metrics Notification channel Operational owner
Install	—	curl -L https://heyclau.de/downloads/skills/windsurf-collaborative-development.zip -o windsurf-collaborative-development.zip && unzip -o windsurf-collaborative-development.zip -d ./windsurf-collaborative-development Copy install	—	—	curl -L https://heyclau.de/downloads/skills/codex-automations-orchestrator-capability-pack.zip -o codex-automations-orchestrator-capability-pack.zip && unzip -o codex-automations-orchestrator-capability-pack.zip -d ./codex-automations-orchestrator-capability-pack Copy install
Config	—	—	—	—	—
Citations	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationcode.claude.com Source methodology →	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationdocs.windsurf.com Package (SHA-256 pinned)/downloads/skills/windsurf-collaborative-development.zip Source methodology →	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationcode.claude.com Submitted by kiannidev2026-06-14 Source methodology →	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Submitted by MkDev112026-06-04 Source methodology →	Source repositorygithub.com 2026-07-18T19:14:44+00:00 Documentationgithub.com Package (SHA-256 pinned)/downloads/skills/codex-automations-orchestrator-capability-pack.zip Source methodology →
Claim	Unclaimed	Unclaimed	Unclaimed	Unclaimed	Unclaimed

Open 4 picks in the interactive comparison tool

1. 01
   Bootstrap CLAUDE.md with /init in Claude Code

   Generate a starting CLAUDE.md with the built-in /init command, then refine it. Manage memory files with /memory, imports, and path-scoped rules.

   Invocation:/init

   Added 8mo ago

   commandsReview firstSource-backedReview first

   Safety ✓ Privacy ✓by JSONbored

   Copy install

   Why it made the cut

   Bootstrap CLAUDE.md with /init in Claude Code is included because it has safety notes present, privacy notes present, source-backed source posture.

   Reach for instead

   If this will touch credentials, local files, or production systems, inspect the upstream source first.

2. 02
   

   Windsurf AI-Native Collaborative Development Skill

   Master collaborative AI-assisted development with Windsurf IDE's Cascade AI, multi-file context awareness, and Flow patterns for team wor...

   Level:advancedType:generalVerified:draft

   Added 9mo ago

   skillsTrustedFirst-partyLow risk

   Safety ✓ Privacy ✓by JSONbored

   Copy install

   Why it made the cut

   Windsurf AI-Native Collaborative Development Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

3. 03
   Compaction and Memory Hygiene for Long Claude Sessions

   Use /compact, /memory, and CLAUDE.md hygiene to keep long Claude Code sessions accurate.

   Added 1mo ago

   guidesReview firstSource-backedReview first

   Safety ✓ Privacy ✓by kiannidev

   Copy install

   Why it made the cut

   Compaction and Memory Hygiene for Long Claude Sessions is included because it has safety notes present, privacy notes present, source-backed source posture.

   Reach for instead

   If this will touch credentials, local files, or production systems, inspect the upstream source first.

4. 04
   Manage Prompt and Context Hygiene in Long Coding Sessions

   Keep long Claude Code sessions focused and reviewable.

   Added 1mo ago

   guidesReview firstSource-backedReview first

   Safety ✓ Privacy ✓by MkDev11

   Copy install

   Why it made the cut

   Manage Prompt and Context Hygiene in Long Coding Sessions is included because it has safety notes present, privacy notes present, source-backed source posture.

   Reach for instead

   If this will touch credentials, local files, or production systems, inspect the upstream source first.

5. 05
   Codex Automations Orchestrator Capability Pack Skill

   Plan and operate recurring agent automations with deterministic prompts, guardrails, and output contracts.

   Level:expertType:capability-packVerified:validated

   Added 3mo ago

   skillsTrustedFirst-partyLow risk

   Safety ✓ Privacy ✓by JSONbored

   Copy install

   Why it made the cut

   Codex Automations Orchestrator Capability Pack Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

6. 06
   MCP Server Security Hardening Skill

   Harden Model Context Protocol servers with practical controls for auth, tool scope, input validation, and supply-chain risk.

   Level:advancedType:generalVerified:draft

   Added 3mo ago

   skillsTrustedFirst-partyLow risk

   Safety ✓ Privacy ✓by JSONbored

   Copy install

   Why it made the cut

   MCP Server Security Hardening Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

7. 07
   V0 Rapid UI Prototyping Workflow Skill

   Turn prompts and screenshots into React components, iterate in v0, then pull the result into your codebase.

   Level:advancedType:generalVerified:draft

   Added 9mo ago

   skillsTrustedFirst-partyLow risk

   Safety ✓ Privacy ✓by JSONbored

   Copy install

   Why it made the cut

   V0 Rapid UI Prototyping Workflow Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

8. 08
   Troubleshooting High CPU Memory And Search Problems In Claude Code

   Fix Claude Code CPU, memory, and search issues with /compact, /heapdump, and ripgrep steps from official docs.

   Added 1mo ago

   guidesReview firstSource-backedReview first

   Safety ✓ Privacy ✓by kiannidev

   Copy install

   Why it made the cut

   Troubleshooting High CPU Memory And Search Problems In Claude Code is included because it has safety notes present, privacy notes present, source-backed source posture.

   Reach for instead

   If this will touch credentials, local files, or production systems, inspect the upstream source first.

9. 09
   Agent Evals Regression Gate Skill

   Create a practical eval harness with golden sets, rubric scoring, and release gates for agentic workflows.

   Level:advancedType:generalVerified:draft

   Added 3mo ago

   skillsTrustedFirst-partyLow risk

   Safety ✓ Privacy ✓by JSONbored

   Copy install

   Why it made the cut

   Agent Evals Regression Gate Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

10. 10
    AI Agent Observability and Incident Response Skill

    Add actionable observability for prompts, tools, latency, failures, and safety events with incident response runbooks.

    Level:advancedType:generalVerified:draft

    Added 3mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    AI Agent Observability and Incident Response Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

11. 11
    AI Search Ranking Content Cluster Strategy Skill

    Content strategy skill for intent mapping, pillar/cluster architecture, internal linking, and structured publishing workflows.

    Level:advancedType:generalVerified:draft

    Added 3mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    AI Search Ranking Content Cluster Strategy Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

12. 12
    Base L2 Smart Contract Launchpad Skill

    Practical launch workflow for Base contracts from local build to verified deployment and post-launch operations.

    Level:advancedType:generalVerified:draft

    Added 3mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    Base L2 Smart Contract Launchpad Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

13. 13
    Browser Agent Workflow Automation Skill

    Design production-ready browser agent workflows with reliable selectors, guardrails, and replayable runbooks.

    Level:advancedType:generalVerified:draft

    Added 3mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    Browser Agent Workflow Automation Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

14. 14
    

    Cursor Windsurf AI Code Editor Skill - Claude Code Skills

    Configure and optimize Cursor and Windsurf AI code editors for maximum productivity.

    Level:advancedType:generalVerified:draft

    Added 8mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    Cursor Windsurf AI Code Editor Skill - Claude Code Skills is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

15. 15
    FastAPI Python API Development Skill

    Build high-performance async REST APIs with FastAPI, Python's fastest-growing web framework.

    Level:advancedType:generalVerified:draft

    Added 8mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    FastAPI Python API Development Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

16. 16
    Git-Cliff Release Changelog Capability Pack Skill

    Generate stable, high-signal changelogs and release notes with git-cliff and conventional commit discipline.

    Level:expertType:capability-packVerified:validated

    Added 3mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    Git-Cliff Release Changelog Capability Pack Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

17. 17
    

    GitHub Actions AI-Powered CI/CD Automation Skill

    Build intelligent CI/CD pipelines with GitHub Actions, AI-assisted workflow generation, automated testing, and deployment orchestration.

    Level:advancedType:generalVerified:draft

    Added 9mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    GitHub Actions AI-Powered CI/CD Automation Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.

18. 18
    

    GitHub Actions Secure CI/CD Capability Pack Skill

    Deep CI/CD security and reliability skill for GitHub Actions with reusable workflows, policy checks, and hardened automation patterns.

    Level:expertType:capability-packVerified:validated

    Added 3mo ago

    skillsTrustedFirst-partyLow risk

    Safety ✓ Privacy ✓by JSONbored

    Copy install

    Why it made the cut

    GitHub Actions Secure CI/CD Capability Pack Skill is included because it has maintainer-built package, safety notes present, privacy notes present, first-party source posture.