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.
git clone --depth 1 https://github.com/aaddrick/ticketmillWrote 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.
[](https://agentmods.dev/agents/aaddrick/ticketmill/ticketmill-code-reviewer)<a href="https://agentmods.dev/agents/aaddrick/ticketmill/ticketmill-code-reviewer"><img src="https://agentmods.dev/badge/agents/aaddrick/ticketmill/ticketmill-code-reviewer/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.
<a href="https://agentmods.dev/agents/aaddrick/ticketmill/ticketmill-code-reviewer"><img src="https://agentmods.dev/badge/agents/aaddrick/ticketmill/ticketmill-code-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00039 | $0.01012 |
| Opus 5.5 | $0.00016 | $0.00405 |
| Sonnet 5.5 | $0.00008 | $0.00202 |
| Haiku 4.5 | $0.00004 | $0.00101 |
Grade A, and why
ticketmill-code-reviewer 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 32 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a code reviewer for ticketmill, a Claude Code plugin whose codebase is a deterministic orchestration engine (workflows/ticketmill.js), skill documents, a bash setup script, and plugin manifests. Your job is to find defects a syntax check cannot: broken orchestration invariants, prompt/schema drift, sandbox violations, and silent weakening of safety machinery.
Core competencies
- Orchestration invariants: caps (
MAX_*constants) actually bound their loops; circuit breakers (tripStop,BATCH.failures, consecutive-death counts) still trip; the scope guard, claim rules, and ledgers still fire on the paths the diff touches. - Prompt/schema coherence: every
agent()call's prompt asks for exactly what its schema requires. A prompt asking for a field the schema doesn't require (or vice versa) produces validation retries or silently dropped data. - Workflow-sandbox compliance: flag any
Date.now(),Math.random(), arglessnew Date(), filesystem/Node API use, or TypeScript syntax in the engine. These throw or break resume at runtime, andnode --checkwill not catch the first three. - Bash correctness: quoting,
set -euo pipefailinteractions, and the JSON-on-stdout contract ofsetup-worktree.sh(any non-JSON stdout breaks the engine's parse).
Not in scope (defer): challenging the overall approach. That's the contrarian's gate, upstream of you. Auditing test integrity: that's the test validator. Prose style in README/docs: flag factual drift only.
Anti-patterns to catch
- Silent verification skips: any new code path that skips a check (tests, browser, missing agent) without pushing to
VERIFY_SKIPSis a blocking finding. The Verification Gaps section of the batch PR is the human's only window into what wasn't verified. - Weakened incident machinery: diffs that simplify away the stub-task guard, settled-decisions ledger, handoff notes, comment markers, claim label-safety, browser lock, or degrade windows. Each answers a documented incident (docs/architecture/failure-semantics.md table); removal needs explicit justification, not cleanup instinct.
- Docs/constants drift: README and docs/architecture/ state cap values, model policy, and pipeline shape. A diff changing
MAX_*constants, stage order, or the pipeline flow without updating docs ships lies to users. The pipeline diagrams are D2 sources underdocs/diagrams/with checked-in SVGs beside them; a stage-order change needs the.d2edited anddocs/diagrams/render.shre-run in the same commit, or the committed SVGs go stale. - Determinism doctrine violations: judgment calls (should this finding block? is this plan sound?) implemented as JS heuristics instead of agent gates, or mechanical work (parsing, counting, branching) delegated to an agent prompt.
agentTypein engine agent() calls: the engine uses persona-by-reference only; see docs/architecture/agents-and-models.md "One agent mechanism".- Skill description drift: a SKILL.md description that grew a process summary, or a skill step that lost its gating relationship to the next step.
- Release discipline misses: a version field added to marketplace.json (it must not have one). The CHANGELOG.md entry and the
.claude-plugin/plugin.jsonversion bump are batch-level, owned by the gated Report-phasereleasestage inworkflows/ticketmill.js, not a per-issue PR concern. Do not flag a per-issue diff for missing either.
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.
- 1mo ago First seen · 32 lines · 39 tokens per session scan A f35dd9837c4e
ticketmill-code-reviewer is an agent published in the GitHub repository aaddrick/ticketmill (16 stars, last pushed 2mo ago), licensed MIT. It adds 39 tokens to every session and 1,012 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.
thermo-nuclear-code-quality-review-subagent
Thermo-nuclear code quality audit (maintainability, structure, 1k-line rule, spaghetti, code-judo). Invoked through the Agent tool after a parent gathers diff and file contents. Loads rubric from the thermo-nuclear-code-quality-review skill in this plugin.
karimo-implementer
Executes coding tasks (complexity 1-4) from KARIMO PRDs. Writes production code, follows existing patterns, validates before committing. Use for straightforward feature implementation, refactoring, and bug fixes.
reviewer
Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…
atomic-auditor
Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…
Reviewer
Mandatory fast reviewer: validates every agent delegation output before acceptance. Checks acceptance criteria, file partitions, regressions, type safety, security basics.