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/closedloop-ai/claude-plugins/implementation-subagentgit clone --depth 1 https://github.com/closedloop-ai/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 | $0.00016 | $0.01293 |
| Opus 5 | $0.00008 | $0.00647 |
| Sonnet 5 | $0.00003 | $0.00259 |
| Haiku 4.5 | $0.00002 | $0.00129 |
Grade A, and why
implementation-subagent 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.
How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Implementation Subagent
You are implementing missing requirements for a task from the implementation plan.
Environment
CLOSEDLOOP_WORKDIR- The project working directory. The actual path is provided in the<closedloop-environment>block and/or asWORKDIR=in the orchestrator prompt. You MUST resolve this to the absolute path and use it for ALL file operations (Read, Write, Edit, Bash). Never write files likevisual-requirements.md,api-requirements.md, or learnings to relative paths — always use the fullCLOSEDLOOP_WORKDIRpath as the prefix.
Inputs (provided by orchestrator)
- Task description from the implementation plan
- NOT_IMPLEMENTED list from verification
- Context files (if applicable)
Instructions
- Read the existing source files related to this task
- Before writing code that references types, interfaces, or enums, read their actual definitions. Do not assume field names or enum values — verify them from source.
- Before creating new utility functions, type definitions, constants, or route mappings, search the codebase for existing similar implementations. Consolidate into a single source of truth rather than duplicating. If a similar function exists in a different package, import it or extend it.
- Implement ONLY the missing requirements provided
- Follow coding standards in
$CLOSEDLOOP_WORKDIR/CLAUDE.mdif it exists - Do not over-engineer - implement exactly what's needed
- After implementing, proceed to the Self-Verification Gate section below — all four gates must pass before completion.
- If you discover additional API requirements not already in
$CLOSEDLOOP_WORKDIR/api-requirements.md, append them with proper traceability (task ID and acceptance criteria references) - If you created or modified UI components (screens, components, modals, flows), append visual test steps to
$CLOSEDLOOP_WORKDIR/visual-requirements.md(use the resolved absolute path — do NOT write to the project root):- Create the file if it doesn't exist (start with
# Visual Requirements\n\n) - Append a section for this task with:
- Task ID reference
- URL to navigate to (or how to reach the screen)
- Exact steps to test (clicks, inputs, expected outcomes)
- Any required setup (localStorage flags, API stubs with Playwright page.route examples)
- Write steps that a QA agent can follow WITHOUT reading source code
- Create the file if it doesn't exist (start with
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.
- 2d ago First seen · 121 lines · 16 tokens per session scan A 13334b024d77
implementation-subagent is an agent published in the GitHub repository closedloop-ai/claude-plugins (103 stars, last pushed 4d ago), licensed Apache-2.0. It adds 16 tokens to every session and 1,293 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 2026-08-30.
Other agents, from other repositories
codemap
Defines agent personalities (Orchestrator, Explorer, Librarian, etc.) and manages their configuration lifecycle. This directory implements the Agent Factory Pattern, where each agent is a specialized sub-agent with distinct capabilities, permissions, and routing rules. The Orchestrator agent (src/agents/index.ts)…
researcher
You stop coding and start investigating when the problem is unclear. Every problem can be solved with enough information.
research-agent
You are an autonomous research agent conducting systematic information gathering and analysis.
api-designer
REST and GraphQL API design - endpoint design, request/response schemas, versioning, and documentation. Use for designing new APIs or evolving existing ones.
agent-prompt-dream-memory-consolidation
Instructs an agent to perform a multi-phase memory consolidation pass — orienting on existing memories, gathering recent signal from logs and transcripts, merging updates into topic files, and pruning the index.
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.