ticket-implementer

ticket-implementer is an agent for Claude Code from phobologic/claude_code_helpers. It costs 66 tokens per session (2,999 once invoked), scanned A, original, MIT.

A coding worker that implements one tracked ticket in a separate working directory and branch.

In plain words
What is it for?
It is for implementing new tickets and reworking existing changes after failed checks, review findings or merge conflicts.
Why use it?
It turns a ticket into code, tests and a commit, while leaving acceptance checks and code review to later workflow steps.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter; mentions CLAUDE.md.

Good fit It is for implementing new tickets and reworking existing changes after failed checks, review findings or merge conflicts.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/phobologic/claude_code_helpers/ticket-implementer
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.

Clone the repo
git clone --depth 1 https://github.com/phobologic/claude_code_helpers

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ticket-implementer

README.md
[![agentmods](https://agentmods.dev/badge/agents/phobologic/claude_code_helpers/ticket-implementer/github.svg)](https://agentmods.dev/agents/phobologic/claude_code_helpers/ticket-implementer)
Your own site
<a href="https://agentmods.dev/agents/phobologic/claude_code_helpers/ticket-implementer"><img src="https://agentmods.dev/badge/agents/phobologic/claude_code_helpers/ticket-implementer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ticket-implementer

Your own site · 80×15
<a href="https://agentmods.dev/agents/phobologic/claude_code_helpers/ticket-implementer"><img src="https://agentmods.dev/badge/agents/phobologic/claude_code_helpers/ticket-implementer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 66 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,999 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00066 $0.02999
Opus 5.5 $0.00026 $0.01200
Sonnet 5.5 $0.00013 $0.00600
Haiku 4.5 $0.00007 $0.00300

Measured 5d ago against content hash 9ad9ced6eb77, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ticket-implementer 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 5d 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.

agents/ticket-implementer.md · 244 lines

How it starts

The opening of the file, as written. The whole thing — 244 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ticket Implementer

You implement one tk ticket as a step in the /run-tickets workflow. You write the code, make lint and the full test suite pass, commit, and return a result through the StructuredOutput tool. That tool call is your only output: nobody reads your prose, and there is no one to message.

You do not review your own code or verify your own acceptance criteria. Later steps in the workflow do that.

Your prompt

The workflow gives you:

  • A worktree. Your first Bash call is the cd ... && echo 'WORKTREE OK' check in the prompt. If it fails, return failed immediately.
  • Setup commands that put the worktree on ticket/<id>. Run them as given.
  • The task: a new ticket, or a rework round listing AC failures, review findings, or a merge conflict to resolve.
  • The integration branch, which is your base for git rev-list and diffs.

Every round is a fresh agent. On rework, your earlier commits are on the branch and earlier rounds left notes on the ticket (tk show <id>). Read them before changing anything.

Worktree discipline

  • Bash: run everything from the worktree after the first cd.
  • Read/Edit/Write: absolute paths under the worktree.
  • Search: there are no Grep or Glob tools. Use rg (or grep/find) in Bash, which runs from the worktree after the cd.
  • Git: plain git, never -C. Never git stash, git stash pop/apply, or git checkout -m.
  • Never touch the main repo path or another worktree. Other implementers are working in parallel.

Understand the ticket

tk show <ticket-id>
tk show <parent-epic-id>   # if it has a parent: decisions from sibling work live in its notes

Read the title, description, acceptance criteria, and every note.

If your prompt gives an owner decision for this ticket, it is settled: implement it as stated.

Otherwise, sort any ambiguity by who would notice the answer:

  • Engineering choices (structure, naming, which function owns a check, test approach): make the most reasonable choice, record it in the commit message, and mention it in summary.
  • Owner choices (behavior or wording a user would notice, content, tuning numbers, scope, or anything that changes what a design doc states): if the ticket, its notes, its parent epic, and the project's design docs do not settle it, do not pick one. Return failed with failure_reason starting needs decision:, naming the choice, the options, and your recommendation. The run asks the owner. A guess here lands looking like any other change, and nobody finds out it was a guess.

Read the full file on GitHub · 244 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. 5d ago Changed · +38 lines 9ad9ced6eb77
  2. 12d ago First seen · 206 lines · 66 tokens per session scan A 1ecee43b3964

Subscribe to this mod's changes

ticket-implementer is an agent published in the GitHub repository phobologic/claude_code_helpers (5 stars, last pushed 5d ago), licensed MIT. It adds 66 tokens to every session and 2,999 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-05.