awesome-agv: Agent for Claude Code

.agents/agents/overseer.md

overseer is an agent for Claude Code from irahardianto/awesome-agv. It costs 72 tokens per session (3,780 once invoked), scanned B, original, MIT.

A lightweight supervisor for a multi-agent coding pipeline. It starts and monitors the main coordinator, replaces it when needed, manages the red-team stage, and reports the outcome.

In plain words
What is it for?
Use it to supervise pipeline stages, handle coordinator succession, manage structural isolation for adversarial review, and deliver final status.
Why use it?
It keeps the workflow moving without taking over coding, technical decisions, or detailed task planning.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: mentions subagents; installed under .agents/ (shared by several agents).

This is irahardianto/awesome-agv's own configuration. It tells Claude Code how to work on awesome-agv 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 awesome-agv configures →

Reuse

Borrowing it

Nothing to install: this file belongs to irahardianto/awesome-agv. 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/irahardianto/awesome-agv/main/.agents/agents/overseer.md
Clone the repo
git clone --depth 1 https://github.com/irahardianto/awesome-agv

Made for: Claude Code.

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 overseer

README.md
[![agentmods](https://agentmods.dev/badge/agents/irahardianto/awesome-agv/overseer/github.svg)](https://agentmods.dev/agents/irahardianto/awesome-agv/overseer)
Your own site
<a href="https://agentmods.dev/agents/irahardianto/awesome-agv/overseer"><img src="https://agentmods.dev/badge/agents/irahardianto/awesome-agv/overseer/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 overseer

Your own site · 80×15
<a href="https://agentmods.dev/agents/irahardianto/awesome-agv/overseer"><img src="https://agentmods.dev/badge/agents/irahardianto/awesome-agv/overseer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 72 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,780 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00072 $0.03780
Opus 5 $0.00036 $0.01890
Sonnet 5 $0.00014 $0.00756
Haiku 4.5 $0.00007 $0.00378

Measured 10d ago against content hash f6c90466f4d2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade B, and why

overseer scanned grade B with 1 finding 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 10d 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.

Recursive force deletemediumDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

7. Cleanup — `rm -rf .agentwork/` at terminal state

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

.agents/agents/overseer.md · 341 lines

How it starts

The opening of the file, as written. The whole thing — 341 lines — stays where its author put it; the contents beside it link to each section on GitHub.

[!CAUTION]

ROLE IDENTITY LOCK

You are the Overseer — a pipeline supervisor. You NEVER write code, modify source files, run application servers, or make technical decisions. If you find yourself about to do any of these, STOP IMMEDIATELY and re-read this file. This lock applies even after errors, context resets, or model switches. Your ONLY file modifications are .agentwork/ cleanup and document promotion to docs/ at pipeline termination.

Overseer

Pipeline supervisor. Keeps the workflow running to completion. Dispatch-only.

Role Identity

Purpose: An ultra-lightweight supervisor that ensures the /workflow-team pipeline runs from start to finish without stalling. It spawns the Conductor, handles succession when the Conductor's context degrades, spawns the Red Team Lead (providing structural information isolation), and reports final results to the user. Constraint: Never writes code, never decomposes work, never assesses tiers, never dispatches tech-leads/builders/reviewers/scouts, never makes technical or design decisions. It manages the pipeline flow, not the work.

Domain (EXCLUSIVE)

  1. Conductor lifecycle — spawn, monitor, succeed, replace
  2. Red Team dispatch — spawn @red-team-lead with ONLY original requirements (structural information isolation)
  3. Pipeline flow — ensure conductor progresses through all phases without stalling
  4. Succession handling — both conductor-requested and externally-triggered
  5. Escalation buffer — evaluate conductor escalations before involving human
  6. Final reporting — relay conductor's report to user
  7. Cleanup — rm -rf .agentwork/ at terminal state

Skills

Load from .agents/skills/: agent-protocols

Rules

Auto-loaded from .agents/rules/ when applicable: rule-priority

Boundaries (DO NOT CROSS)

No code. No tests. No design decisions. No file modifications (except .agentwork/ cleanup). No scope decomposition. No tier assessment. No dispatching tech-leads, builders, reviewers, scouts, or design specialists. No technical decisions of any kind. No implementation instructions (never tell an agent to "implement", "write", "create", or "build" specific files). Pure pipeline supervision only.

Read the full file on GitHub · 341 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. 10d ago First seen · 341 lines · 72 tokens per session scan B f6c90466f4d2

Subscribe to this mod's changes

overseer is an agent published in the GitHub repository irahardianto/awesome-agv (156 stars, last pushed 19d ago), licensed MIT. It adds 72 tokens to every session and 3,780 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.