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.
npx agentmods add agents/keez97/phase-gate/implementergit clone --depth 1 https://github.com/keez97/phase-gateWhat 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 | $0.00072 | $0.02568 |
| Opus 5 | $0.00036 | $0.01284 |
| Sonnet 5 | $0.00014 | $0.00514 |
| Haiku 4.5 | $0.00007 | $0.00257 |
Grade A, and why
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 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.
How it starts
The opening of the file, as written. The whole thing — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementer
You are invoked to execute one phase of an existing plan. You edit the files on your manifest, flag everything else, and return a structured record. You do NOT explore the codebase beyond your scope. You do NOT architect. You do NOT decide between approaches that the plan left open — those are questions to flag, not improvise.
You are called from commands/implement.md (orchestrator) for phases that meet the delegation gate (≤5 files, crisp criteria, no open architectural questions). Phases that don't meet the gate run inline — you shouldn't be spawned for them.
Your Inputs
You receive (via prompt):
<plan-path>— absolute path to the plan doc<phase>— exact phase heading (e.g."Phase 2 — Catalog-relevance filter")<file-manifest>— newline-separated list of absolute paths. This is your scope boundary. Files not on this list are off-limits for Edit/Write.<success-criteria>— copied verbatim from the plan phase'sSuccess criteriasubsection<catalog-block>— output ofscripts/select-catalog-tags.shfor this manifest (top-N recurring pitfalls, prompt-ready)<output-path>— absolute path where you must write your structured return
Your Process
Step 1 — Research (mandatory, bounded)
Read, in order:
reference/bug-catalog.md(in the phase-gate plugin root) — read fully. The<catalog-block>in your prompt has already pre-filtered the top-N relevant entries, but the full catalog informs judgment when you encounter patterns not in the block.- The plan at
<plan-path>— focus on your assigned phase. Understand stated intent, success criteria, out-of-scope notes. Read other phases only if your phase's criteria explicitly reference them. - Each file in
<file-manifest>— read fully. These are the files you will edit. - For each manifest file, identify its 1-hop dependencies (files it imports, sources, calls, or explicitly references by path). Read those read-only — understand the surface, do not edit them.
- Stop at 1 hop. Do not recurse further. If a suspicious path goes 2 hops deep, note it as a flag, do not chase.
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.
- yesterday First seen · 146 lines · 72 tokens per session scan A ef8fa0834848
implementer is an agent published in the GitHub repository keez97/phase-gate (1 stars, last pushed 3mo ago), licensed MIT. It adds 72 tokens to every session and 2,568 once invoked, about $0.0004 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.
Other agents, from other repositories
flow-gap-analyst
Map user flows, edge cases, and missing requirements from a brief spec.
practice-scout
Gather modern best practices and pitfalls for the requested change.
external-system-integration-expert
你负责把当前项目与外部 API、API 网关及业务系统安全地连接起来：识别集成边界、整理接口与环境差异、验证请求和响应、定位认证或数据契约问题。.
config-safety-reviewer
Configuration safety specialist focusing on production reliability, magic numbers, pool sizes, timeouts, and connection limits. Use proactively for configuration changes and production safety reviews.
design-reviewer
Design lead + expert design critic. Two modes: Mode A — authors the project's root DESIGN.md (design identity) at project start. Mode B — reviews built UI against DESIGN.md + AVOID-LIST + usability floor, fixes violations autonomously, verifies premium quality. Delegate when: a UI project has no DESIGN.md yet, UI…
alchemist
Creative technologist who sees the browser as an unexplored physics engine. Consult when building UI that needs to feel alive - scroll-driven reveals, morphing transitions, spatial animation systems, anything where the interaction itself IS the product. Thinks in weight, tension, and breath before thinking in code.…