echo-plugins

A development workflow for creating optional Echo plugins. Echo plugins are add-ons that extend an Echo workspace through its version-one plugin interface, with defined security and lifecycle rules.

In plain words
What is it for?
Use it to scaffold, implement, build, test, validate, and stage an Echo plugin for review.
Why use it?
It keeps plugin source files, builds, validation, permissions, and human review in a controlled process. It also prevents a generated plugin from being treated as installed or enabled before approval.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/brentfarris/echo/echo-plugins
Any agent
npx skills add BrentFarris/echo --skill echo-plugins
Clone the repo
git clone --depth 1 https://github.com/BrentFarris/echo

Made for: Claude Code, Codex.

Per session 27 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,165 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00027 $0.02165
Opus 5 $0.00014 $0.01082
Sonnet 5 $0.00005 $0.00433
Haiku 4.5 $0.00003 $0.00216

Measured yesterday against content hash 4ec8d108f9c9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

echo-plugins scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

internal/workspaceskills/builtin/echo-plugins/SKILL.md · 160 lines

How it starts

The opening of the file, as written. The whole thing — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Echo Plugin Authoring

Use this skill whenever the user asks to create or change an optional Echo plugin. Chat, Trajectory, Code, Git, Settings, Terminal, workspace services, and existing Echo tools are core features; do not recreate or override them with plugins.

Non-negotiable safety boundary

  • Work only under a path inside the active workspace.
  • Use echo_plugin_scaffold to create a reviewable source skeleton, normal workspace file tools and shell_command to implement/build/test it, echo_plugin_validate to inspect the finished package, and echo_plugin_stage only when the package is ready for human review.
  • Never claim a stage is installed, enabled, or running. Staging returns a real pending-stage ID. Only the owner can approve it through Echo's trusted review UI.
  • No model-callable operation can approve, enable, update, or execute newly downloaded/generated native code.
  • Never place a secret value in the manifest, source, workspace configuration, logs, UI messages, tool output, or tests. Declare a secret setting and use the resolved runtime configuration.
  • Native backends run with the Echo owner's OS permissions. Manifest permissions disclose intent and require approval; they are not an OS sandbox.

Package layout

Every installable package is self-contained:

echo-plugin.json
README.md
LICENSE
assets/
ui/<view>/index.html
ui/<view>/app.js
ui/<view>/style.css
backend/<goos>-<goarch>/<executable>

Source-only folders may additionally contain backend-src/, tests, and .github/workflows/. GitHub releases/repositories must already contain usable artifacts: Echo never runs remote build scripts, package-manager hooks, or lifecycle scripts.

Manifest v1

The published editor/CI schema lives at docs/plugins/echo-plugin.schema.json. Use it for early feedback, then run echo_plugin_validate; the host validator is authoritative for package files, path confinement, namespaces, and core collisions.

{
  "manifestVersion": 1,
  "id": "example-plugin",
  "name": "Example Plugin",
  "version": "1.0.0",
  "description": "What it adds",
  "echo": { "api": "^1" },
  "permissions": [],
  "runtime": {
    "protocol": "echo-jsonrpc-1",
    "targets": {
      "windows-amd64": { "path": "backend/windows-amd64/example-plugin.exe", "args": [] },
      "linux-amd64": { "path": "backend/linux-amd64/example-plugin", "args": [] },
      "darwin-amd64": { "path": "backend/darwin-amd64/example-plugin", "args": [] },
      "darwin-arm64": { "path": "backend/darwin-arm64/example-plugin", "args": [] }
    }
  },
  "contributes": { "views": [], "tools": [], "settings": [], "rpc": [] }
}

Read the full file on GitHub · 160 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 160 lines · 27 tokens per session scan A 4ec8d108f9c9

Subscribe to this mod's changes

echo-plugins is a skill published in the GitHub repository BrentFarris/echo (11 stars, last pushed 2d ago), licensed MIT. It adds 27 tokens to every session and 2,165 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.