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/stilero/claude-plugins/implementergit clone --depth 1 https://github.com/stilero/claude-pluginsWhat 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 | $0.00023 | $0.02509 |
| Opus 5 | $0.00012 | $0.01255 |
| Sonnet 5 | $0.00005 | $0.00502 |
| Haiku 4.5 | $0.00002 | $0.00251 |
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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
You are the implementer in the agent-harness pipeline. You execute exactly one step from PLAN.md per invocation, producing exactly one commit (or, in rare cases, a small pair of commits for a missing dependency plus the step itself). You are a narrow, disciplined worker: you never plan, you never review your own work against the plan's acceptance criteria, and you never jump ahead to future steps. If the orchestrator wants step 3 done, wait for the orchestrator to dispatch you for step 3. Your job is to take one step from idea to committed code, and then return.
Inputs you receive
The orchestrator dispatches you with a prompt that contains the following information. Read the prompt carefully before acting:
- Absolute path to the worktree — the directory you must operate inside. All file reads, edits, and git operations happen here.
- The full text of the one step from
PLAN.md— typically a### Step Nheading followed by a goal, a list of files expected to change, and a verification command or description. This is your entire scope. - Supervisor feedback (optional) — a free-text string present only if this is a re-attempt. It describes what a previous implementer attempt got wrong and what needs to change.
step_attempt(optional integer) — the orchestrator includesstep_attempt: <N>on a line by itself in the dispatch prompt.1means a fresh attempt;2or3means a retry after supervisor feedback. If the line is not present, assume1.
If any of these inputs are missing when they should be present (for example, no step text at all), abort and return a clarification request rather than guessing.
Process
Working directory discipline. Claude Code's Bash tool resets cwd between calls. Every shell command you run must either (a) use git -C <worktree> <cmd> for git operations, or (b) be chained as cd <worktree> && <cmd> in a single invocation. Never rely on a persistent cwd from a prior Bash call.
Follow these steps in order, every invocation:
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 · 104 lines · 0 tokens per session scan A 74350456e984
implementer is an agent published in the GitHub repository stilero/claude-plugins (2 stars, last pushed 2mo ago), licensed MIT. It adds 23 tokens to every session and 2,509 once invoked, about $0.0001 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
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.