eddo-work

A structured workflow for implementing complex software tasks through phases such as refining the task, coding, reviewing, completing, and creating a GitHub pull request. It works with a task-tracking tool called eddo-todo.

In plain words
What is it for?
Moving a task from clarification through implementation and review, tracking changed files, updating task metadata, and preparing a pull request.
Why use it?
It provides checkpoints and checklists so complex work can be organized, reviewed, and recorded consistently.

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/walterra/eddoapp/eddo-work
Any agent
npx skills add walterra/eddoapp --skill eddo-work
Clone the repo
git clone --depth 1 https://github.com/walterra/eddoapp

Made for: Claude Code, Codex.

Per session 68 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,715 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 $0.00068 $0.02715
Opus 5 $0.00034 $0.01358
Sonnet 5 $0.00014 $0.00543
Haiku 4.5 $0.00007 $0.00271

Measured 2d ago against content hash 4b56899c0b11, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

eddo-work 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 2d ago.

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.

packages/chat-agent/skills/eddo-work/SKILL.md · 382 lines

How it starts

The opening of the file, as written. The whole thing — 382 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Eddo Work

Structured implementation phases for complex tasks. Uses eddo-todo for task management.

  • eddo-todo: Core task management commands (REQUIRED - load this skill first)

Prerequisites

Load the eddo-todo skill first and set up variables:

# Inside Docker container:
EDDO="${PI_CODING_AGENT_DIR:-/home/agent/.pi/agent}/skills/eddo-todo/eddo-todo.js"
BRANCH=$(git branch --show-current 2>/dev/null || echo "none")
METADATA='{"agent:session":"YOUR_PI_SESSION_ID","agent:model":"YOUR_PI_MODEL","agent:cwd":"YOUR_PI_CWD","agent:branch":"'$BRANCH'"}'

File Tracking

Track all files you modify during implementation in the agent:files metadata field (array of relative paths). This enables the Eddo UI to show which files were touched per task.

Update metadata when you modify files:

# After modifying files, update the parent todo's metadata
$EDDO update $PARENT_ID -M '{"agent:session":"...","agent:files":["src/Button.tsx","src/utils.ts"]}' -m "Updated file list"

Get current file list from git:

# Files changed in current branch vs main
git diff --name-only main...HEAD

Trigger Pattern

User: "what next?"
Agent: Suggests task from `$EDDO next`
User: "let's work on this" / "start working"
Agent: Enters work mode with structured phases

Critical Rules

  • Test every change - never assume fixes work without verification
  • Only complete subtasks when FULLY WORKING and VERIFIED
  • Never git commit/push without user confirmation
  • Follow all workflow phases - no shortcuts
  • STOP at gates for user check-in
  • MANDATORY: Execute reflection checklist at end of EVERY phase
  • MANDATORY: Output reflection results with ✅/❌ for each item

Phase 1: REFINE

Start session:

$EDDO start <id> -M "$METADATA" -m "Starting REFINE phase"
$EDDO get <id>

Investigate:

  • Review todo description for context and requirements
  • Read relevant codebase files
  • Ask clarifying questions if needed
  • Define success criteria
  • Log surprising findings: $EDDO note <id> "Discovered X depends on deprecated Y"

Read the full file on GitHub · 382 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. 2d ago First seen · 382 lines · 68 tokens per session scan A 4b56899c0b11

Subscribe to this mod's changes

eddo-work is a skill published in the GitHub repository walterra/eddoapp (58 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 68 tokens to every session and 2,715 once invoked, about $0.0003 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.