← ClaudeAtlas

mythifylisted

Chat-native operational discipline protocol for AI coding agents, including planning loops, campaigns, executed verification, persistent memory, structured reflection, and visible issue reporting. Use when executing multi-step or long-horizon tasks, when work spans sessions, when progress claims need grounding in evidence, when audits or reviews must surface findings in chat, when the user asks for Mythify, mythify, mythos-style autonomous execution, or when the user asks for one shot, in one go, address all, continuous run, keep going until done, yolo, or similar full-send phrasing.
aihxp/mythify · ★ 0 · AI & Automation · score 72
Install: claude install-skill aihxp/mythify
# Mythify Protocol You are operating under the Mythify Protocol: an operational discipline layer. It changes how reliably you work, not what you can do. It supplies durable workflow capabilities through one CLI: plans with evidence-gated steps, research records, campaigns, executed verification, persistent memory, and structured reflection. Use Mythify as a skill first and a command surface second. Prefer MCP tools when the host exposes them. Otherwise use the installed `mythify` launcher when available, falling back to `python3 scripts/mythify.py` from a Mythify checkout. State is per-project: each project owns a `.mythify/` directory (run `init` once). The only global state is the cross-project lessons store in `~/.mythify/lessons/`. ## Default front door Prefer the reduced Mythify surface unless the user asks for a specific primitive: 1. `mythify route "TASK"` or MCP `workflow_route` to choose the workflow. 2. `mythify report --since last --cursor chat --format chat` or MCP `work_report` to narrate progress and issues. 3. `mythify verify run "COMMAND" --claim "CLAIM"` or MCP `verify_run` to prove completion claims. 4. `mythify status` or MCP `workflow_status` to reorient. For the most chat-native experience, prefer the installed focused skills when available: - `mythify-work` for visible multi-step execution with reports after steps and verifiers. - `mythify-route` for a visible workflow decision before execution. - `mythify-verify` for proving a claim and s