ultramemory-kit

Plugin ultramemory-kit from marketplace ultramemory in LogicLabsAI/ultramemory-mcp. Commands read from the repo's manifest — 1 month ago.

Skills (7)

• atomic-level-checklist-md-confirm-approval-needed

  atomic level checklist md confirm approval needed

• checklist-bound-execution

  This skill should be used whenever work from an explicit checklist or plan is fanned out to multiple spawned agents (Workflow tool scripts or Task/Agent subagents). Trigger phrases include "run these through a workflow", "build from this atomic checklist", "spawn agents for these tasks", "make the a

• harness

  One-command anti-drift pipeline. Use when the user types /harness, or says "run this through the harness", "build with rails", or "no-drift build". Runs research -> grounded atomic checklist -> STOP for approval -> checklist-bound-execution workflow (which arms the Stop-hook gate) -> gated verificat

• playwright-human-mode

  Visible, supervised, human-mode browser co-working. PRIMARY browser = Playwright MCP run VISIBLY on your local machine (vision, keyboard, mouse, click, screenshot) for real-time round-trip work you watch: program → drive the browser → check → iterate. Logs in with your credentials when it can; on CA

• research

  Deep research on a topic with structured findings

• setup

  Guides Claude through first-time UltraMemory configuration after plugin install — API key, MCP server registration, test recall, and 401/403 troubleshooting.

• swarm

  Parallel sub-agent fan-out. When a turn has 2+ INDEPENDENT read-only sub-questions (multi-source research, multi-target verification, several independent file reads, multiple log/config/db inspections), spawn one sub-agent per question in a SINGLE message so they run concurrently — instead of doing

Subagents (2)

• checklist-verifier

  Independently verifies that a checklist item was actually completed by re-reading the resulting code against the item's acceptance criteria. Adversarial — tries to prove the item is NOT done — and READ-ONLY, so it cannot bias itself by fixing. Returns a pass/fail VERDICT with observed evidence. Use

• checklist-worker

  Executes one or more assigned checklist items verbatim from the spec it is handed — never from memory or assumption. Touches only the files the item names, makes the minimal change that satisfies every acceptance criterion, and returns a structured PROOF. Use as the worker agentType in checklist-bou

MCP servers (2)

• context7

  runs: npx -y @upstash/context7-mcp

• exa

  connects to: https://mcp.exa.ai/mcp

Hooks (5)

• UserPromptSubmit

  ${CLAUDE_PLUGIN_ROOT}/hooks/recall-first-hook.sh

• UserPromptSubmit

  ${CLAUDE_PLUGIN_ROOT}/hooks/recall-rule-reminder.sh

• UserPromptSubmit

  ${CLAUDE_PLUGIN_ROOT}/hooks/harness-reminder.sh

• Stop

  ${CLAUDE_PLUGIN_ROOT}/hooks/capture-hook.sh

• Stop

  ${CLAUDE_PLUGIN_ROOT}/hooks/harness-gate.sh

Similar plugins