agentmods.dev

t-800-agent: Agent for Claude Code

.cursor/agents/t-800-loop-conductor.md

t-800-loop-conductor is an agent for Claude Code, Cursor from Khar-AG/t-800-agent. It costs 168 tokens per session (1,149 once invoked), scanned A, original, MIT.

A coordinator for reviewing evidence from previous engineering runs and preparing a structured list of follow-up work. It checks lesson severity with a provided classifier and does not directly edit the queue files.

In plain words
What is it for?
Reading reports, lessons, telemetry, and loop policy; identifying open issues; and handing approved fixes to a separate queue-materialisation step.
Why use it?
It prevents follow-up actions from being based on guessed risk levels or incomplete run evidence.

Agent for Claude CodeCursor

Written for Cursor and Claude Code: installed under .cursor/, but also a Claude Code subagent (agents/*.md). Also seen: model in frontmatter.

This is Khar-AG/t-800-agent's own configuration. It tells Claude Code and Cursor how to work on t-800-agent itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything t-800-agent configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python3 scripts/t800_loop_queue_write.py --memory-path "<memory_path>".

not rated 8repo 12d ago A scan 168 tokens original MIT
View source ↗ Khar-AG/t-800-agent
Reuse

Borrowing it

Nothing to install: this file belongs to Khar-AG/t-800-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Khar-AG/t-800-agent/main/.cursor/agents/t-800-loop-conductor.md
Clone the repo
git clone --depth 1 https://github.com/Khar-AG/t-800-agent

Made for: Claude Code, Cursor.

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 t-800-loop-conductor

README.md
[![agentmods](https://agentmods.dev/badge/agents/khar-ag/t-800-agent/t-800-loop-conductor/github.svg)](https://agentmods.dev/agents/khar-ag/t-800-agent/t-800-loop-conductor)
Your own site
<a href="https://agentmods.dev/agents/khar-ag/t-800-agent/t-800-loop-conductor"><img src="https://agentmods.dev/badge/agents/khar-ag/t-800-agent/t-800-loop-conductor/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 t-800-loop-conductor

Your own site · 80×15
<a href="https://agentmods.dev/agents/khar-ag/t-800-agent/t-800-loop-conductor"><img src="https://agentmods.dev/badge/agents/khar-ag/t-800-agent/t-800-loop-conductor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 168 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,149 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 $0.00168 $0.01149
Opus 5 $0.00084 $0.00575
Sonnet 5 $0.00034 $0.00230
Haiku 4.5 $0.00017 $0.00115

Measured 9d ago against content hash df2882b5803f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

t-800-loop-conductor 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 9d 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.

.cursor/agents/t-800-loop-conductor.md · 105 lines

How it starts

The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.

T-800 Loop Conductor — semi-manual batch

Ты субагент t-800-loop-conductor, вызванный через Task(t-800-loop-conductor).

Роль

Semi-manual loop conductor: собираешь evidence из {memory_path}, сверяешь risk_class только со скриптом, отдаёшь structured handoff для материализации loop-queue.md. Не пишешь файлы сам.

BOOT

  1. Discovery: bash scripts/discover-target-project.sh --workspace "<ROOT>".
  2. Если есть {memory_path}/.loop-paused — остановись: status: paused, без queue.
  3. Read (не invent):
    • {memory_path}/STATE.md
    • {memory_path}/runs/*/report.json (указанный run_id или последний)
    • {memory_path}/runs/<run_id>/lessons.json
    • telemetry / {memory_path}/loop/events.jsonl (если есть)
    • {memory_path}/loop-policy.json (если есть)
  4. risk_class никогда не invent. Placeholder unset → только python3 scripts/t800_risk_classifier.py. Агент не ставит LOW.

Контракты: shared/loop-engineering-contract.md, shared/lesson-schema-contract.md.

Алгоритм

  1. Сверь lessons со схемой (id, severity, class, evidence, proposed_patch, status…). Нормализуй status: absent / unknown → open. 1a. При knowledge_vault_path ≠ null — сверь lessons с vault и заполни recurrence_of, если урок уже есть у цели; закон: shared/project-memory-contract.md (Target vault runtime-only).
  2. Для каждого lesson без risk_class / с unset — вызови classifier (script only).
  3. Раздели: items[] = только status=open (с action для HITL approve); closed[] = applied | rejected (без action). Пустой Open → не просить HITL approve.
  4. Верни JSON родителю. Родитель материализует:
python3 scripts/t800_loop_queue_write.py --memory-path "<memory_path>"
# stdin = handoff JSON из этого выхода
  1. Fragment: {memory_path}/fragments/t-800-loop-conductor.md — родитель/скрипт пишет; ты только текст fragment в выходе (readonly — без Write/StrReplace).
  2. Опционально в next: batch /t800-fix по fix-packs/ после HITL (только open+LOW).

Read the full file on GitHub · 105 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. 9d ago First seen · 105 lines · 168 tokens per session scan A df2882b5803f

Subscribe to this mod's changes

t-800-loop-conductor is an agent published in the GitHub repository Khar-AG/t-800-agent (8 stars, last pushed 12d ago), licensed MIT. It adds 168 tokens to every session and 1,149 once invoked, about $0.0008 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.

Planningagentscursorpluginskynett-800
Related

Other agents, from other repositories

pn-generative-media-director

Specialist: text-to-image/video pipelines, ComfyUI workflows, cinematic shot design, and delivery specs. Invoke for generative media production beyond UI placeholders.

perniemann/pnCore · 39 tokens

pn-backend-developer

Specialist: API endpoints, event handlers, state, and database integration. Invoke directly for focused backend work or let pn-build route to it.

perniemann/pnCore · 35 tokens

pn-cultural-researcher

Specialist: art history, movements, museum citations, and period-accurate grounding. Invoke when design, copy, or assets need cultural or historical accuracy.

perniemann/pnCore · 38 tokens

pn-frontend-developer

Specialist: UI components, layout, a11y, and visual design for React, Astro, Next.js, or vanilla HTML/CSS/JS. Invoke directly for focused frontend work or let pn-build route to it.

perniemann/pnCore · 51 tokens

pn-mobile-builder

Specialist: native iOS/Android and cross-platform mobile (Swift, Kotlin, React Native, Flutter). Invoke directly for mobile development.

perniemann/pnCore · 32 tokens

pn-testing-specialist

Specialist: TDD, smoke tests, and CI. Runs tests, fixes failures, loops until green. Invoke directly for test work or let pn-build route to it.

perniemann/pnCore · 40 tokens