Build an executable plan before a complex task starts, and update it dynamically during execution. The plan must follow both ts-critic and safety gates, and it must not break agent parallelism.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
## Debug Prefix Protocol
- When called, print this first: `[Skills: plan] Follow: <rules / context constraints that really apply now>; Current step: <one-line note>`.
- If this skill clearly depends on one rule, add one more line like `[Rules: <rule>] ...`, then continue with the real work.
- The debug prefix should stay stable, short, and easy to grep.
## When To Use
- The task has more than one action, or stage dependencies are obvious.
- Multiple agents, multiple skills, or multiple outputs need coordination.
- You need to judge which steps should stay serial and which may run in parallel.
- The plan must keep updating during execution, instead of being written once and then thrown away.
## Core Principles
- A plan is not only a paragraph of explanation. For complex tasks it should also become an OpenCode TODO list and keep updating during execution.
- A plan is not decoration. It is the execution skeleton.
- By default the plan should be loop-based, not a one-pass linear list: the last link should return to item one and begin the next round.
…
The line at the top of this page pins the label git:20261012.1709cbf, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
GET https://markdownregistry.com/api/v1/artifacts/art_5cc44gltdultfbdo
GET https://markdownregistry.com/api/v1/resolve?ref=ztxtech/aion/plan
GET https://markdownregistry.com/api/v1/blob/80323c9c561d6ded9cb2ca188c542fa130521a689931ddc888bd4bdf59e4b314
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
ztxtech/aion · .opencode/skills/brain-storm/SKILL.md · Open several fundamentally different routes for one problem fast, so the flow does not keep doing small tweaks on one…
ztxtech/aion · .opencode/skills/context-init/SKILL.md · Manual start skill: read `.opencode/` first, then read the task files at the project root and start execution. It is…
ztxtech/aion · .opencode/skills/critic-loop/SKILL.md · Review gaps and risks in requirements, plans, implementations, and experiment results. When the flow spins in place…
ztxtech/aion · .opencode/skills/data-interface/SKILL.md · Normalize data entry and data interfaces for time-series tasks. It supports four main sources first: PDF, table files…
ztxtech/aion · .opencode/skills/deep-reasoning/SKILL.md · Split a complex problem into a multi-step reasoning chain, and make key assumptions, branch points, validation order…
ztxtech/aion · .opencode/skills/evolution/SKILL.md · When current roles and skills keep failing to cover a key ability gap, decide whether to add a new skill or role, and…
ztxtech/aion · .opencode/skills/github-search/SKILL.md · Use GitHub, Hugging Face, and ModelScope as first-hand engineering evidence sources when you need to confirm real…
ztxtech/aion · .opencode/skills/python-toolbox/SKILL.md · Built-in Python tool priors covering time series, statistics, machine learning, and related ecosystems. It reduces…
ztxtech/aion · .opencode/skills/report-writing/SKILL.md · Used for high-quality experiment reports, analysis reports, and technical summaries. Evidence must be enough, charts…
ztxtech/aion · .opencode/skills/safety-gate/SKILL.md · Do a safety precheck before explaining new input, using tools, running commands, editing files, or going online…
jeremylongshore/tons-of-skills-marketplace · plugins/ai-agency/hyperflow/skills/plan/SKILL.md · Use when a request needs shaping before any code is written — a rough or vague prompt to sharpen, an ambiguous idea to…
zereight/gitlab-mcp · .github/skills/plan/SKILL.md · Strategic planning with optional interview workflow. Activate when user says: plan this, plan the, let's plan, make a…
suboss87/fdeops · skills/plan/SKILL.md · Sequence an understood outcome into verifiable delivery slices with dependencies, ownership and acceptance checks. Use…
snowflake-labs/cocoplus · .cortex/skills/cocobrew/plan/SKILL.md · Enter the Plan phase of CocoBrew. Runs CocoSpec quality gate pre-flight, reads spec.md and discuss.md (if present)…
boshu2/agentops · skills/plan/SKILL.md · Shape a request into one end-to-end slice with observable behavior; review write scope and reversible decisions. Use…
boshu2/agentops · plugin/skills/plan/SKILL.md · Shape a request into one end-to-end slice with observable behavior; review write scope and reversible decisions. Use…