codifying-tasks
Turns recurring work into reusable skills. Use when a task being done now has been done before in this project, when the user says "make this repeatable", "turn this into a skill", "we'll be doing this again", or at the natural end of a multi-step task that will clearly recur. The rule of thumb - do it once by hand, codify it the second time.
correcting-mistakes
Self-corrects skill and instruction files after a genuine instruction defect is confirmed. Use when Claude errors during skill execution and finds the instruction was at fault, or when the user reports a mistake AND verification confirms the instruction (not the execution) caused it. Applies succinct fixes so a real defect never repeats — without mangling correct instructions on a false alarm.
creating-skills
Creates high-quality Claude skills following official best practices. Use when the user asks to create a new skill, improve an existing skill, or needs guidance on skill authoring. Includes guided scoping, proper structure, naming conventions, progressive disclosure patterns, current frontmatter options, and quality validation.
analyzing-source
Conducts in-depth analysis of a specific source or topic, producing a comprehensive, information-dense summary for research synthesis. Use when detailed analysis and documentation of an individual source is needed as part of a larger research effort. Primarily used by the research-analyst agent inside the deep-research workflow.
deep-research
Conducts comprehensive multi-source research on a topic - clarifies scope, plans research angles, spawns parallel analyst subagents to study each source in depth, and synthesizes findings into cited documentation. Use when the user wants thorough research, a literature review, a competitive scan, or says "deep research", "research this properly", "build me a research doc".
auditing-access
Read-only audit of what Claude Code can currently see and do - settings layers, permission rules and modes, connected MCP servers, hooks, additionalDirectories, installed plugins, and sensitive files not covered by deny rules. Use when the user asks "what can you access", "audit my setup", "is anything exposed", or before starting work with confidential material. Makes no changes; never echoes secret or credential values.
securing-claude
Hardens Claude Code permissions with safe defaults for knowledge work. Use when the user wants to protect secrets or sensitive files, control what runs without asking, set up guardrails before working with confidential material, or says "secure my setup", "lock this down", "set up permissions". Writes settings.json rules the harness enforces - stronger than any prompt instruction.
capturing-knowledge
Captures durable learnings into a second-brain knowledge base (a folder with a conventions.md or KB-declaring CLAUDE.md) when conditions warrant - a reusable insight, a framework crystallizing, a constraint discovered, a decision trail completed. Triggers on "save this to my notes/knowledge base", "file that under", "worth keeping", or the natural pause at the end of substantive work where a real learning emerged. Only operates when a knowledge base exists; not for process chatter, debugging logs, or tactical one-offs.
managing-daily-notes
Manages daily notes in a second-brain markdown knowledge base (a folder with a conventions.md or KB-declaring CLAUDE.md) - finding, creating, reading, and updating them, including task carryover with provenance. Trigger on "my daily note", "create today's daily note", "carry over my tasks", "close out the day" when working in such a knowledge base.
organizing-notes
Keeps a markdown knowledge base consistent - frontmatter on every note, controlled tags, links between related notes and entities. Use whenever creating or editing notes in a knowledge base, and when asked to clean up, normalize, or reorganize notes. Reads the knowledge base's conventions.md as the authority.
processing-inbox
Processes and routes files from a second-brain knowledge base's inbox folder (e.g. _inbox/) into their proper homes per the base's conventions.md. Use when asked to check, clean, process, sort, or organize the knowledge-base inbox. Categorizes by content analysis, moves what's clear, asks about what isn't, never deletes.
setting-up-knowledge-base
Initializes a plain-markdown knowledge base Claude can maintain - folder structure, a conventions file, and a CLAUDE.md so every future session knows the rules. Use when the user wants to start a second brain, notes vault, or knowledge base, or wants Claude to take over maintaining an existing folder of notes. Works with Obsidian or any markdown editor.
analyzing-source
Conducts in-depth analysis of a specific source or topic, producing comprehensive summaries for research synthesis. Use when you need detailed analysis and documentation of individual sources as part of a larger research effort.
correcting-mistakes
Self-corrects skill and command instructions after resolving errors. Use when Claude errors during skill/command execution and finds a solution, or when user indicates a mistake was made. Finds the relevant instruction file and applies succinct fixes.
creating-skills
Creates high-quality Claude skills following official best practices. Use when the user asks to create a new skill, improve an existing skill, or needs guidance on skill authoring. Includes proper structure, naming conventions, progressive disclosure patterns, and quality validation.