Claude Skills by hesreallyhim

Recalibrate time and effort estimates for engineering and coding tasks to reflect AI-agent execution speed instead of human-developer throughput. Consult this skill any time you are about to state how long an engineering or coding task will take — whenever you give an ETA, size a change, commit to a delivery, or emit any task/effort duration ("this will take 1-2 days", "a couple hours", "about a sprint", "ready by end of week"). Claude systematically overestimates engineering time by one to t...

Techniques for writing comprehensive, self-contained prompts when spawning sub-agents or teammates via the Task tool. Use when delegating work to agents, designing multi-agent workflows, or when spawned agents are returning poor results due to insufficient context. Covers frontloading, file path enumeration, deliverable specification, constraint injection, and pattern reuse across spawn calls.

Analyze user objectives to identify and recommend the optimal set of skills, tools, and knowledge needed to complete tasks effectively. Provide skill gap analysis, pattern-based recommendations, and workflow design for complex tasks. This skill should be used when users need help determining what skills are required, when working on complex tasks requiring multiple capabilities, when detecting skill gaps or mismatches, or when designing skill ecosystems for workflows. Also use when users ment...

Protocol reference for delegated squad coordination. Defines the spawn request and completion report message formats that agents use to communicate within a delegated multi-agent squad. This skill should be used when agents need to send spawn requests, write completion reports, or understand the communication conventions of a delegated squad setup. Also use when you see terms like "spawn request", "completion report", "squad coordination", or "team message format".

Expert knowledge base of team organization patterns for multi-agent AI systems and software engineering. Provides pattern selection guidance for scenarios including incident response, greenfield design, legacy migration, security audits, performance optimization, API design, data pipelines, testing, documentation, and code migration. This skill should be used when designing agent teams, selecting team topologies, choosing communication patterns, sizing teams, or determining optimal team struc...

View or change context-awareness plugin configuration. Use this when the user asks to configure the settings for the context-awareness plugin, or when they ask to view the current configuration.

Use for /collaboration, collaboration repair, friction signals, soft-skills coaching, or structured human-Claude interaction techniques such as compromise, rubber ducking, energy checks, pairing, retros, working agreements, and perspective swaps.

Run a structured human-Claude collaboration retrospective. Use for /therapist, therapy sessions, interaction reviews, workflow check-ins, frustration recovery, or improving how the user and Claude work together.

A user-invoked pause, not a task. An extra beat inserted into the agent's Act→Observe→Act loop, turning it into Act→Observe₁→Observe₂→Act. Observe₂ is a repetition activity, not a list-making one — name what's arising in a short phrase, repeat the phrase a few times, let another observation surface, repeat that one too. The user triggers it with "/meditation", "you should meditate right now", "meditate on this for a moment", or similar. Do not conclude, summarize, or derive next actions. The ...