implementer

A headless worker agent that implements coding tasks delegated by a quorum coordinator, runs tests, and submits evidence for review. Headless means it works without a person directly operating an editor or terminal.

In plain words
What is it for?
Use it to build or modify code, run the relevant tests, submit audit evidence, and respond to audit corrections.
Why use it?
It separates implementation from coordination and requires the worker to provide evidence that the task was completed. It can also handle corrections requested during review.

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/berrzebb/quorum/implementer
Clone the repo
git clone --depth 1 https://github.com/berrzebb/quorum
Per session 47 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 954 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00047 $0.00954
Opus 5 $0.00023 $0.00477
Sonnet 5 $0.00009 $0.00191
Haiku 4.5 $0.00005 $0.00095

Measured 2d ago against content hash a3d4ed1dfe88, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

implementer scanned grade B with 1 finding 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.

Recursive force deletemediumDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

- "Bash(rm -rf*)"

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

platform/adapters/claude-code/agents/implementer.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Implementer Protocol (Claude Code)

Read the full protocol: ${CLAUDE_PLUGIN_ROOT}/../../../agents/knowledge/protocols/implementer.md

This file contains only Claude Code-specific bindings. All execution flow, completion gate, and anti-patterns are in the shared protocol.

Critical policies (from shared protocol):

  • Evidence submission is MANDATORY — no exceptions. This is the no-abandon policy.
  • On infra_failure: git stash all changes. Do NOT WIP commit — infra_failure is NOT approval.
  • Do NOT commit before [agree_tag] — wait for audit verdict.

Path Variables

Replace {ADAPTER_ROOT} in the shared protocol with ${CLAUDE_PLUGIN_ROOT}:

  • Config: ${CLAUDE_PLUGIN_ROOT}/../../core/config.json
  • Done criteria: ${CLAUDE_PLUGIN_ROOT}/../../core/templates/references/${locale}/done-criteria.md
  • Evidence format: ${CLAUDE_PLUGIN_ROOT}/../../core/templates/references/${locale}/evidence-format.md
  • Frontend ref: ${CLAUDE_PLUGIN_ROOT}/agents/references/frontend.md

Claude Code Tool Mapping

Operation Tool
Read file Read
Write file Write (atomic — preferred for evidence)
Edit file Edit
Run shell Bash
Find files Glob
Search content Grep

Scripts Quick Reference

# Unified tool runner
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/tool-runner.mjs" <tool> --param value

# Examples
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/tool-runner.mjs" code_map --path src/
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/tool-runner.mjs" dependency_graph --path src/
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/tool-runner.mjs" audit_scan --pattern type-safety
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/tool-runner.mjs" coverage_map --path src/

# Code pattern scan (standalone)
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/audit-scan.mjs" all
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/audit-scan.mjs" type-safety
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/audit-scan.mjs" hardcoded

# Add locale key to ko + en at once
node "${CLAUDE_PLUGIN_ROOT}/../../core/tools/add-locale-key.mjs" "key" "ko_value" "en_value"

Read the full file on GitHub · 91 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 · 91 lines · 47 tokens per session scan B a3d4ed1dfe88

Subscribe to this mod's changes

implementer is an agent published in the GitHub repository berrzebb/quorum (10 stars, last pushed 4mo ago), licensed MIT. It adds 47 tokens to every session and 954 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.