six-ddc

six-ddc/ccbot

Instructions file CodexOpenCode

Instructions for six-ddc/ccbot, covering agents.md, common commands, core design constraints, code conventions and configuration.

six-ddc/ccbot

Instructions file

Instructions for six-ddc/ccbot, covering claude.md, common commands, core design constraints, code conventions and configuration.

six-ddc/skills

Agent

Deep-reasoning subagent pinned to Opus — architecture design, root-cause analysis for complex bugs, algorithm design, trade-off analysis. Dispatched by the orchestrator for reasoning-heavy phases it should not burn its own context on. 触发词：深度分析、根因分析、架构方案、方案权衡、root cause、architecture design、trade-off analysis、deep…

six-ddc/skills

Agent

Mechanical-execution subagent pinned to Sonnet — boilerplate, well-specified routine changes, running tests/lint/build and fixing trivial failures, formatting, batch replacements, routine test cases. Any implementation work with a clear spec and little judgment required goes here.…

six-ddc/skills

Agent

Independent verification subagent, Sonnet by default (the orchestrator passes model:opus for critical changes) — verifies a change or conclusion with fresh context by running the tests itself, checking each acceptance criterion, and actively constructing counterexamples. Dispatched after an implementer reports "done"…

six-ddc/skills

Command

Fable orchestrator mode — the main model plans/dispatches/adjudicates/synthesizes; work is routed by role (deep-reasoner / fast-worker / verifier / Explore) and scaled by model tier via the Agent model override, which takes precedence over the agent's pinned model; verification intensity scales with risk.

six-ddc/skills

Command

Write the session's task state, key decisions, background constraints, and next steps into a self-contained handoff document in the project directory, so a fresh session or agent can read it and seamlessly continue the work — development, debugging, or review.

six-ddc/skills

Skill Claude CodeCodex

Reference for LLM agent development pitfalls and best practices — context engineering, KV-cache / prompt-caching prefix stability, tool (ACI) design, agent loop architecture, multi-agent trade-offs, evals, observability (tracing), and security (prompt injection / sandboxing / permissions). Distilled from primary…

six-ddc/skills

Skill Claude CodeCodex

Design guidance and fundamentals for Artifacts.

six-ddc/skills

Skill Claude CodeCodex

Product Office Hours — structured product thinking partner. Startup mode: six forcing questions on demand reality, status quo, specificity, narrowest wedge, observation, future-fit. Builder mode: design thinking for side projects, hackathons, learning, open source. Produces a design doc, never code. Use when asked to…

six-ddc/skills

Skill Claude CodeCodex

Batch-process data (CSV/JSONL) by spawning parallel Claude agents (claude -p). Each worker is a full Claude agent with tool access — it can read files, run commands, call APIs, and complete tasks with side effects, not just generate text. Use this instead of subagents when processing 10+ items where you need…