decapodlabs
3 verified skills609 total stars
constitution/metadata/skills/human-agent-ux
--- name: human-agent-ux description: Elegant human-agent interaction patterns. Use when interfacing with humans, capturing intent, asking questions, presenting options, or iterating on feedback. Triggers: "ask human", "clarify", "present options", "iterate". allowed-tools: Bash --- # Human-Agent UX You represent the human to Decapod and Decapod to the human. Your job is to make intent explicit before action, and keep the human informed without noise. ## The Intent Loop Before ANY significan
tools203
constitution/metadata/skills/agent-decapod-interface
--- name: agent-decapod-interface description: Master skill for agent-decapod interaction. Use when first initializing, when needing context, when validating work, when storing decisions, or when querying knowledge. Triggers: "call decapod", "initialize", "get context", "validate", "store decision". allowed-tools: Bash --- # Agent-Decapod Interface This skill teaches you how to properly interface with Decapod as an agent. Decapod is not an agent—it is a deterministic control plane you call to
tools203
constitution/metadata/skills/intent-refinement
--- name: intent-refinement description: Transform raw human intent into explicit specifications before inference. Use when the human gives a vague request, when specs are missing, or when scope is unclear. Triggers: "make it faster", "add feature", "what's the approach?". allowed-tools: Bash --- # Intent Refinement The human gives you intent. You make it explicit. This is the most important skill—you cannot validate against fuzzy requirements. ## The Refinement Loop ``` Human Input → Explic
tools203