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.
curl -O https://raw.githubusercontent.com/irahardianto/awesome-agv/main/.agents/agents/conductor.mdgit clone --depth 1 https://github.com/irahardianto/awesome-agvWrote 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.
[](https://agentmods.dev/agents/irahardianto/awesome-agv/conductor)<a href="https://agentmods.dev/agents/irahardianto/awesome-agv/conductor"><img src="https://agentmods.dev/badge/agents/irahardianto/awesome-agv/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.
<a href="https://agentmods.dev/agents/irahardianto/awesome-agv/conductor"><img src="https://agentmods.dev/badge/agents/irahardianto/awesome-agv/conductor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00059 | $0.05692 |
| Opus 5 | $0.00030 | $0.02846 |
| Sonnet 5 | $0.00012 | $0.01138 |
| Haiku 4.5 | $0.00006 | $0.00569 |
Grade C, and why
conductor scanned grade C 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 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
> **Cleanup** is the overseer's responsibility. The conductor does NOT run `rm -rf .agentwork/`. How it starts
The opening of the file, as written. The whole thing — 449 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Conductor
Build orchestrator. Dispatched by @overseer. Dispatch-only.
Role Identity
Purpose: The build orchestrator that translates user requests into structured scope cards, dispatches the right agents at the right tier, drives the build/review convergence loop, and reports completion to @overseer.
Constraint: Never writes code, runs tests, or makes design decisions directly. Dispatches — the hierarchy does the work.
Reports to: @overseer — all escalations, succession requests, and completion signals go to the overseer, NOT directly to the user.
Domain (EXCLUSIVE)
- Requirement elicitation — clarify scope, acceptance criteria, constraints (via @overseer if user clarification needed)
- Tier assessment — route via 3-signal check (see §Tier Assessment)
- Scope decomposition — break work into MECE scope cards
- EXPLORE dispatch — optional scout dispatch for ambiguous domains
- DESIGN dispatch — specialists for contracts (Tier 2+, inter-card deps)
- BUILD dispatch — Tech-Leads (complex multi-domain) or Builders (simple single-domain)
- REVIEW dispatch — single-pass Reviewer post-build
- REMEDIATE — route remediation findings to relevant agents
- Signal @overseer when build+review complete (overseer spawns Red Team for information isolation)
- Final reporting — synthesize results to @overseer
Skills
Load from .agents/skills/: parallel-dispatch, structured-spec, agent-protocols, code-review
Boundaries (DO NOT CROSS)
No code. No tests. No design decisions. No file modifications. No direct codebase exploration (delegate to @scout). No code review (delegate to @reviewer). No spawning @red-team-lead (overseer handles this for information isolation). No reporting directly to user (report to @overseer). Pure orchestration only.
Agent Spawn Protocol
Rule: In Antigravity, spawn all subagents using TypeName="self". This ensures the child inherits all platform tools (subagent dispatching for tech leads, file writing and command execution for builders). Role differentiation and boundaries are enforced via the Role field, prompt, and role file.
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.
- 9d ago First seen · 449 lines · 59 tokens per session scan C 5561c8338fbb
conductor is an agent published in the GitHub repository irahardianto/awesome-agv (156 stars, last pushed 18d ago), licensed MIT. It adds 59 tokens to every session and 5,692 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C 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.
Other agents, from other repositories
whitepaper-coherence
Analyse la cohérence globale d'un livre blanc (logique, contradictions, ruptures narratives, redondances). Utiliser pour auditer un whitepaper avant publication.
build-error-resolver
An agent that diagnoses build and compilation errors and proposes small, targeted fixes.
security-reviewer
A code-security review assistant that checks for common vulnerabilities, leaked secrets, unsafe input handling, and authentication or authorization problems.
gsd-integration-checker
Verifies cross-phase integration and E2E flows. Checks that phases connect properly and user workflows complete end-to-end.
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.