protocol

A set of repository rules for keeping the UI, command-line tools, and agents aligned around the same Markdown-based work records.

In plain words
What is it for?
It is for coordinating development work in a repository that uses Workfile cards, claims, shared knowledge, and changelog entries.
Why use it?
It explains how to find related work, claim files before editing, avoid overlapping changes, and record decisions or blockers.

Agent

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 agents/illodev/workfile/protocol
Clone the repo
git clone --depth 1 https://github.com/illodev/workfile
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 731 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 $0.00000 $0.00731
Opus 5 $0.00000 $0.00365
Sonnet 5 $0.00000 $0.00146
Haiku 4.5 $0.00000 $0.00073

Measured yesterday against content hash 16dfb302e7e9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

protocol 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.

.project/agents/protocol.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Repository operating protocol

This repository uses Repository Workfile schema v2. Repository Markdown files are canonical. The UI, CLI and every agent adapter must use the same services and rules.

Before working

  1. Search related work and knowledge with pnpm workfile search.
  2. Read the card and its relationship neighborhood before substantial code changes.
  3. Claim the card before touching its scope: pnpm workfile card claim ID --scope path,path. Your identity resolves on its own and pnpm workfile agents whoami prints it. Pass --actor only to claim on someone else's behalf: an actor invented by hand does not match the one the edit guard sees.
  4. Inspect active claims and overlapping scopes. Do not overwrite another actor's work.
  5. Load the smallest relevant context; do not inject all workfile memory into every prompt.

While working

  • Keep the card current when scope, state or blockers change.
  • Create cards in the same session for actionable pending work you discover.
  • Record decisions, incidents, conventions or learnings when they change future behavior.
  • Add a changelog fragment for user-visible changes or whenever project policy requires one.
  • Prefer the CLI or an official adapter for mutations; do not hand-edit frontmatter except in an emergency.
  • Never store credentials, tokens or unnecessary sensitive data in Work, Docs, History or Memory.

Work states

  • backlog: identified without a commitment.
  • next: prioritized for the next batch.
  • doing: actively worked and claimed.
  • review: implementation finished, awaiting verification, deployment or approval.
  • blocked: externally blocked; record why.
  • deferred: deliberately postponed; record why.
  • done: verified in an environment where it actually runs. A commit or merge is insufficient.
  • discarded: will not be done; record why.

Read the full file on GitHub · 62 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 · 62 lines · 0 tokens per session scan A 16dfb302e7e9

Subscribe to this mod's changes

protocol is an agent published in the GitHub repository illodev/workfile (2 stars, last pushed 5d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 731 tokens. 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 2026-08-31.