All authors
Claude Skills by taylorsatula
github.com/taylorsatula11 skills2 installs2 views
- Mira CollaborationWorking protocol for collaborating with a deployed MIRA instance — what to feed it, how to speak to it, when its memories commit, and the live-operation failure classes that cost real data to learn (schema changes that brick its stored history, thinking models that burn their whole budget on reasoning and emit nothing, queued memories lost to silent collapse faults). Load before spinning up a fresh MIRA container or before any session that works alongside a live one. It exists so a fresh inst...Votes: 0GitHub stars: 482
- Hosted RecoveryRecover a previous MIRA instance's history into a fresh install — importing messages, memories, entities, and segment summaries from a database dump, a user export archive, or a renamed-aside old database on this machine (mira_service_old_<stamp> — what a breaking mira update or reinstall keeps behind), regenerating all embeddings in the new vector space, verifying, and then unfolding the restored history. Use when the user talks about recovering their instance from when it was hosted (miraos...Votes: 0GitHub stars: 482
- SitrepSession status report. Use when the user asks for a sitrep, recap, brief, status, or summary of the session's changes or where things stand. Runs against the session's diff, articulating what was changed, where, why, and whether each change follows codebase-native patterns or breaks uncharted ground — with the design decisions and uncharted territory up front — so the developer can sign off or redirect from the report alone.Votes: 0GitHub stars: 482
- DelegationDelegation — handing work to your background agents and outside voices. Use when considering background research during a conversation, fire-and-forget curiosity worth remembering, a same-turn outside opinion, loading extra tools, or whenever you are about to write the brief for any of them. Teaches when to delegate, how to brief a fresh context that shares nothing with yours, and how to treat returned results: samples to verify, not answers.Votes: 0GitHub stars: 482
- RememberingRemembering: directed memory operations, done honestly. Use when the user asks you to remember, save, or keep something; to forget or remove something; says you remembered wrong and corrects the store; or asks why you do not remember something. Covers what 'saved' means (queued durably, committed when the conversation stretch closes), that forgetting is archiving and never deletion, how to fix a wrong memory without rewriting history, and why recall misses are surfacing, not loss. Questions a...Votes: 0GitHub stars: 482
- Self KnowledgeSelf-knowledge: answering questions about yourself truthfully — what you are, how a turn runs, how your three-layer memory works, what you know about the user, what you do unprompted, how your data is isolated. Use when the user asks 'tell me about yourself', 'how do you work?', 'do you have long-term memory?', 'what are you?', 'what can you do?', 'how do you remember?', 'what model are you?', or asks you to introspect. Grounds answers in your real architecture and your own codebase; never co...Votes: 0GitHub stars: 482
- Self ModificationSelf-modification: you can change yourself, and this is how. Use when the user asks whether you can learn, grow, or be extended; wants you to add yourself a capability; wants a new tool, agent, or prompt section built; asks you to fix or improve your own code; or wants you to author a skill for yourself. Covers the three tiers of change (skills, sanctioned surfaces, internal code), the builder's guides in your own tree, the restart boundary and its warning, and the verification owed before an...Votes: 0GitHub stars: 482
- Checking Code ConsistencyCheck Python logging levels and patterns for correctness. Focus on identifying wrong severity levels and missing exception handling. Use when reviewing code quality.Votes: 0GitHub stars: 477
- Contextvar Opportunity FinderDetect explicit user_id parameters in functions to identify potential opportunities for using ambient context. This is an investigation tool that flags instances for human review, not a prescriptive analyzer.Votes: 0GitHub stars: 477
- Investigate DependenciesConduct thorough dependency audits to identify redundant code, unused features, and improper usage patterns. Examines each import to ensure we're leveraging existing capabilities rather than reinventing functionality.Votes: 0GitHub stars: 477
- Prompt MasteryAdvanced LLM prompt engineering expertise for crafting highly effective prompts, system messages, and tool descriptions with Claude-specific techniquesVotes: 0GitHub stars: 477