mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Drive a loomcycle agentic runtime from inside Claude Code.
Plugin loomcycle from marketplace loomcycle in denn-gubsky/claude-code-plugin-loomcycle. Commands read from the repo's manifest — 13 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use and it starts local MCP servers. The exact commands are listed below; read them before installing.
loomcycle-configureConfigure a loomcycle runtime — providers, model tiers and aliases, model kinds (chat / decision / embedder) and decision models, user tiers, fallbacks, per-agent sampling, context retention and compaction, agent hooks and hook definitions, interactive runs and review holds, environment variables, d
loomcycle-diff-agentdefsDiff two loomcycle AgentDef versions by def_id and show what changed in system_prompt, tools, model routing, grants, hooks and other fields. Use when the user wants to compare two versions of an agent definition.
loomcycle-import-claude-codeGuide the user through importing a .claude/ directory into loomcycle yaml using the `loomcycle import claude-code` CLI — dry-run first, review the report, then write. Use when the user wants to move Claude Code agents/skills/MCP servers into a loomcycle config.
loomcycle-memoryOperate loomcycle's memory over MCP — the four scopes, semantic recall vs vector search vs key/value, the ASYNCHRONOUS add path and background consolidation, per-scope SQL, the bi-temporal entity graph (valid_at/invalid_at, supersede-not-delete), chunked-graph Documents and the Path VFS on top of th
loomcycle-replay-failed-runWalk through replaying a failed loomcycle run — inspect the failure, adjust the prompt or agent, and re-spawn. Use when a loomcycle run errored or produced a bad result and the user wants to retry with a fix.
loomcycle-spawn-evaluatorSpawn a loomcycle evaluator agent to score the current Claude Code transcript (or a named run) and report the result. Use when the user wants an independent quality/score judgement of work just done via loomcycle.
/loomcycle:accessAsk what loomcycle's access-control rules decide (check, explain, simulate), and manage the groups, rule lists and bindings that make them.
/loomcycle:cancelCancel a running loomcycle agent by agent_id (cascades to sub-agents).
/loomcycle:compactCompact a parked loomcycle run's conversation — summarize its history to free context, then continue.
/loomcycle:connectSet the active loomcycle identity (user id, per-run bearer, base URL) for this Claude Code session so subsequent /loomcycle commands reuse it.
/loomcycle:decideAsk a loomcycle decision model typed questions (choice, yes/no, score) about text you supply, and get each answer with its probabilities.
/loomcycle:documentsWork on whole sets of loomcycle documents — copy or move them between scopes with their relations, take a scope-wide snapshot of a corpus, or classify a document against the tenant ontology.
/loomcycle:erasureReport what loomcycle holds about one subject, or erase it — three tiers, dry-run by default, with an honest account of what it cannot reach.
/loomcycle:evalSubmit an evaluation score (and optional rationale) against a completed loomcycle run.
/loomcycle:fanoutFan out N concurrent loomcycle runs of one agent in a single call and render the combined results, or start them detached.
/loomcycle:filesInspect, extract, keep or import the files a loomcycle run's tools returned (the run inbox), without their bytes passing through the conversation.
/loomcycle:kb-reviewReview proposed knowledge-base relations (statements) in bulk — see what the queue holds, accept or reject whole groups with a dry run first, apply the tenant's review policy, and undo a ruling.
/loomcycle:memoryRead and write loomcycle memory over MCP — semantic recall, key/value, per-scope SQL, and the consolidation queue.
/loomcycle:operator-tokenMint, rotate, retire, or inspect loomcycle OperatorTokenDef bearer tokens — per-principal multi-tenant auth.
/loomcycle:retuneChange a running loomcycle agent's settings without sending it a message — move it to another model, raise its iteration bound, park it for steering, or hold its answer for review.
/loomcycle:reviewRule on a loomcycle run whose answer is held for review — approve it, send it back with feedback, or reject it.
/loomcycle:runSpawn a loomcycle agent run against a registered agent and stream the result into Claude Code.
/loomcycle:runsList recent loomcycle runs for a user, or every run of one team walk, rendered as a table.
/loomcycle:snapshotCapture, list, restore, or delete loomcycle runtime snapshots from inside the IDE.
/loomcycle:steerSend a steering message to a live interactive run — push operator text into a parked or running interactive agent session.
/loomcycle:subjectsMerge two duplicate knowledge-base subjects into one, or bind a subject to the user account it stands for — operator actions on the tenant ontology.
/loomcycle:whoamiShow who the plugin's loomcycle token signs in as — tenant, subject, status, groups, sign-in gate and the access-control lists that apply.
loomcycleruns: ${user_config.bin_path} mcp --upstream ${user_config.base_url}
"${CLAUDE_PLUGIN_ROOT}/hooks/scripts/capture-run-telemetry.sh"
"${CLAUDE_PLUGIN_ROOT}/hooks/scripts/auto-snapshot-on-error.sh"
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Official Next.js skills: adopt and optimize Cache Components, adopt Partial Prefetching, and verify runtime behavior against a running dev server.
6 skills
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.
3 skills · 6 commands · 1 MCP server · hooks
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.