Skills, agent instructions and declarative marketplace definitions for Claude Code, Codex, Gemini, Pi, OpenCode, RedCode and Hermes.
Overview
Skills, agent instructions and declarative marketplace definitions for Claude Code, Codex, Gemini, Pi, OpenCode, RedCode and Hermes. Runtime software lives in reddb-io/redskilled: MCP servers, daemon, apps, tunnel, IDE integrations, hooks, installers and all package publishing. Existing npm package names remain unchanged. Install the exact runtime version declared in runtime.toon using Redskilled's scripts/install-runtime.sh , then register this marketplace in your host. The red-skills-* commands must be on the host's PATH. Hooks and MCP declarations invoke installed commands; they do not download software or execute repository scripts. Edit plugins/ /skills, references, agent instructions or marketplace metadata. This repository contains no executable implementation. Read CONTRIBUTING.md for validation and coordinated changes. The repositories have independent versions. runtime.toon pins the supported runtime; Redskilled's skills.lock.
README
RedSkills
Skills, agent instructions and declarative marketplace definitions for Claude Code, Codex, Gemini, Pi, OpenCode, RedCode and Hermes.
Runtime software lives in reddb-io/redskilled: MCP servers, daemon, apps, tunnel, IDE integrations, hooks, installers and all package publishing. Existing npm package names remain unchanged.
Install
Install the exact runtime version declared in runtime.toon using Redskilled’s
scripts/install-runtime.sh , then register this marketplace in your host.
The red-skills-* commands must be on the host’s PATH. Hooks and MCP declarations
invoke installed commands; they do not download software or execute repository scripts.
Contribute
Edit plugins//skills, references, agent instructions or marketplace
metadata. This repository contains no executable implementation. Read
CONTRIBUTING.md for validation and coordinated changes.
The repositories have independent versions. runtime.toon pins the supported runtime;
Redskilled’s skills.lock.toon pins the content revision used in its distribution.
Historical releases stay available here; new software releases are published by
Redskilled. See the separation contract.
What Ships
Three plugins install from one marketplace. One source tree, several hosts: Claude Code, Codex, Gemini CLI, OpenCode, Pi, GitHub Actions, MCP servers, and the herdr dashboard are generated from or read the same plugin definitions and runtime apps.
| Plugin | Job | Use it when |
|---|---|---|
dev |
Engineering workflow: issue triage, autonomous execution, worktree safety, review-gated landing, process dashboards, runner policy, and codebase orientation. | You want an agent to plan, execute, review, or ship code work. |
memory |
Governed operational memory: decisions, validations, reasoning attempts, stale-claim checks, context packs, and skill telemetry evidence. | You want agents to stop repeating old mistakes after context resets. |
brain |
Project-local knowledge: typed artifacts, personal/project facts, sources, graph connections, search, cited answers, and dashboards. | You want durable knowledge the human may ask about later. |
Maintainer-only plugin:
| Plugin | Job | Use it when |
|---|---|---|
internal |
Maintainer-only skills for operating this repository. Installable through the normal marketplace flow, but active only when plugins.internal.enabled: true is present. |
You maintain red-skills itself. |
Dev
dev owns the engineering workflow: issue pipeline, autonomous execution,
interactive landing, process visibility, setup/adoption checks, codebase
orientation, and three MCP servers — redskilled,
navigator, and rsp.
Core responsibilities:
- Bootstrap RedSkills with
red-setup. - Maintain issue state with
triage,to-spec, andto-tickets. - Execute delegable work with
afk, or dispatch one concrete demand withgo. - Land a hand-worked branch with
retake(the retiredshipmigrated there — ADR 0081). - Resolve human gates with
hitlor safe retries withretake.
Dev guard rails:
- When
plugins.dev.enabled: true, the dev PreToolUse proxy enforces the worktree boundary for agent shell commands:git worktree addmust target a registered lane under.red/tmp/, and branch-moving commands in the primary checkout (git switch,git checkout,git checkout -b,git switch -c,gh pr checkout) are blocked. Create branches throughgit worktree add .red/tmp/worktrees/manual/ -b ...instead. plugins.dev.lock.primary-branchremains the explicit branch-lock workflow flag that setup writes for compatibility and base-pinning integrations.- The dev shell-command guard is controlled by
command_guardin.red/config.yaml. It runs from the agentPreToolUsehook, stays inert unlessplugins.dev.enabled: true, and blocks matching shell commands before execution. These repo-defined rules are additional to the built-in.red/tmpworktree boundary above.globalrules apply everywhere,mainrules apply in the primary session scope (not specifically the Git branch namedmain), andworktreerules apply in/afkand/goworktrees under.red/tmp/. Deny rules supportregex:,prefix:,suffix:,exact:, andglob:. Bare entries with*,?, or[are Bash globs; other bare entries match the exact command, a command prefix, or a command suffix at a shell-command boundary.command_guard.denyremains a legacy alias forcommand_guard.global.
Example policy, not a default:
command_guard:
global:
- "rm -Rf /*"
- "git stash"
- sudo
- 'regex:(^|[;&|[:space:]])curl[[:space:]].*\|[[:space:]]*sh'
main:
- "git rebase"
- "git checkout -b"
worktree:
- "git clean"
plugins:
dev:
enabled: true
Memory
memory is governed operational memory for code agents. Memory is evidence,
not vibes. It stores work evidence that can make future agents safer and
faster: decisions, root causes, validation evidence, reasoning attempts,
stale-claim checks, readiness, handoffs, and skill telemetry evidence — so a
future agent can verify an old claim before acting on it.
Memory modes:
| Mode | Storage | Best for |
|---|---|---|
markdown-only |
Plain notes under .red/memory/notes/. |
Low-risk rollout with explicit store/recall only. |
graph |
Project-local RedDB store at .red/memory/graph.rdb. |
Governed recall, provenance, supersession, readiness, context packs, Workbench, MCP/HTTP, and Skill telemetry. |
Memory is not the Personal-fact store. Personal facts belong in Brain, not Memory. Broad human-facing project knowledge belongs in Brain too.
Start with plugins/memory/README.md.
Brain
brain is a project-local knowledge repository under .red/brain/*. It stores
typed artifacts and graph connections for later search and cited synthesis.
Use Brain for:
- Personal facts, durable preferences, identity context, and relationship notes.
- Project notes, decisions, ideas, questions, sources, bookmarks, and meeting residue.
brain thinkcited answers with confidence and missing-evidence signals.brain dashboardlocal summaries and KPI-style views.- Optional outbound channel actions through the Brain channel bridge.
Start with plugins/brain/README.md.
Recommended Tools
Try a different keyword or remove a filter.
Install
npx skillfish add reddb-io/red-skills