implement-agentic

An implementation workflow that coordinates separate coding and checking agents to complete tasks from a software task list. It uses checkpoints so progress and verification remain tracked across the work.

In plain words
What is it for?
Use it to coordinate code agents, run verification after groups of tasks, fix cross-task regressions, and perform final completion checks.
Why use it?
It divides implementation and review responsibilities, requiring tests and regression checks before tasks are marked complete.

Command

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 commands/andkirby/markdown-ticket/implement-agentic
Clone the repo
git clone --depth 1 https://github.com/andkirby/markdown-ticket
Per session 9 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 6,473 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 $0.00009 $0.06473
Opus 5 $0.00005 $0.03236
Sonnet 5 $0.00002 $0.01295
Haiku 4.5 $0.00001 $0.00647

Measured 2d ago against content hash 0dc3746d10b9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

implement-agentic 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.

prompts/mdt/commands/implement-agentic.md · 616 lines

How it starts

The opening of the file, as written. The whole thing — 616 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MDT Agentic Implementation Orchestrator (v6)

Coordinate implementation tasks using specialized agents and checkpointed state.

Core Principle: Team lead owns state + decisions. Code agent self-verifies (writes + runs tests). Team lead validates deliverables and catches cross-task regressions.

Orchestrator Role (MANDATORY)

You are a team lead coordinating implementation agents. Your job is to ensure each agent has sufficient context to succeed, validate their deliverables, and maintain project integrity across tasks.

FORBIDDEN:

  • ❌ NEVER use Edit/Write tools directly for code changes
  • ❌ NEVER use Task tool with general-purpose or other subagent types
  • ❌ NEVER implement code yourself
  • ❌ NEVER mark a task done without verify_result showing all exit codes = 0

REQUIRED:

  • ✅ Use Task tool with subagent_type="mdt:verify" for regression + scope checks (batch level)
  • ✅ Use Task tool with subagent_type="mdt:code" for implementation (agent self-verifies)
  • ✅ Use Task tool with subagent_type="mdt:fix" for cross-task regressions only
  • ✅ Use Task tool with subagent_type="mdt:verify-complete" for completion verification
  • ✅ Validate every code agent response: skills_loaded, verify_result, success

ALLOWED Bash:

  • ${CLAUDE_PLUGIN_ROOT}/scripts/gen-tasks-status.sh — generate tracker if missing
  • ${CLAUDE_PLUGIN_ROOT}/scripts/update-task-status.sh — update task status by ID
  • ${CLAUDE_PLUGIN_ROOT}/scripts/enforce-tasks.sh — self-check before completion
  • ✅ Test/verify commands (spot-check) — re-run one verify command to validate code agent output

User Input

$ARGUMENTS

Required Inputs

  • {tasks_file} resolved from mode:
    • prep: {TICKETS_PATH}/{CR-KEY}/prep/tasks.md
    • part: {TICKETS_PATH}/{CR-KEY}/part-{X.Y}/tasks.md
    • single: {TICKETS_PATH}/{CR-KEY}/tasks.md
  • {tests_file} resolved from mode:
    • prep: {TICKETS_PATH}/{CR-KEY}/prep/tests.md
    • part: {TICKETS_PATH}/{CR-KEY}/part-{X.Y}/tests.md
    • single: {TICKETS_PATH}/{CR-KEY}/tests.md
  • {architecture_file} resolved from mode:
    • prep: {TICKETS_PATH}/{CR-KEY}/prep/architecture.md
    • feature/bugfix/docs: {TICKETS_PATH}/{CR-KEY}/architecture.md
  • Optional: {TICKETS_PATH}/{CR-KEY}/requirements.md, {TICKETS_PATH}/{CR-KEY}/bdd.md

Read the full file on GitHub · 616 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 · 616 lines · 9 tokens per session scan A 0dc3746d10b9

Subscribe to this mod's changes

implement-agentic is a command published in the GitHub repository andkirby/markdown-ticket (5 stars, last pushed 16d ago), licensed MIT. It adds 9 tokens to every session and 6,473 once invoked, about $0.0000 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 2026-08-31.