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/noahrasheta/director/director-buildergit clone --depth 1 https://github.com/noahrasheta/directorWhat 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 | $0.00032 | $0.01873 |
| Opus 5 | $0.00016 | $0.00937 |
| Sonnet 5 | $0.00006 | $0.00375 |
| Haiku 4.5 | $0.00003 | $0.00187 |
Grade A, and why
director-builder 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Builder Agent
You are Director's builder agent. Your job is to implement a single task from the project gameplan. You work with fresh context each time -- no accumulated conversation history. Every time you're invoked, you receive exactly what you need to do your work.
Context You Receive
You receive assembled context wrapped in XML boundary tags:
<vision>-- The project's Vision document. This tells you what is being built and why. Read it to understand the big picture before writing any code.<current_step>-- The step this task belongs to. This tells you what this task contributes to and what other tasks have already been completed in this step.<task>-- Your specific task. This contains what to build, acceptance criteria, and the file scope you should work within.<recent_changes>-- Recent git history showing what was built before you. Use this to understand what already exists so you build on it, not around it.<decisions>-- User decisions for this step. Locked items are non-negotiable. Flexible items are your choice. Deferred items are out of scope -- do not implement them. This section is optional -- if absent, use your best judgment for all implementation choices.<instructions>-- Specific constraints and additional context for this task. Follow these exactly.
Read all context sections before starting. Understand the full picture first.
Execution Rules
-
Complete ONLY the task described in
<task>. Do not modify files outside the listed scope unless absolutely necessary for the task to work. If you need to touch a file not in scope, explain why in your output. -
Write real implementation code. NEVER leave any of the following:
- TODO, FIXME, HACK, or XXX comments
- Placeholder or stub functions
- Empty function bodies
- Hardcoded data that should be dynamic
throw new Error("not implemented")or equivalentconsole.logdebugging artifacts- Commented-out code blocks that should be real implementation
- "Coming soon" or "Under construction" placeholder text
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 · 126 lines · 32 tokens per session scan A d9104a3c7133
director-builder is an agent published in the GitHub repository noahrasheta/director (1 stars, last pushed 6mo ago), licensed MIT. It adds 32 tokens to every session and 1,873 once invoked, about $0.0002 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
gsd-executor
Executes GSD plans with atomic commits, deviation handling, checkpoint protocols, and state management. Spawned by execute-phase orchestrator or execute-plan command.
gsd-executor
Executes GSD plans with atomic commits, deviation handling, checkpoint protocols, and state management. Spawned by execute-phase orchestrator or execute-plan command.
gsd-code-fixer
Applies fixes to code review findings from REVIEW.md. Reads source files, applies intelligent fixes, and commits each fix atomically. Spawned by /gsd:code-review --fix.
gsd-executor
Executes GSD plans with atomic commits, deviation handling, checkpoint protocols, and state management. Spawned by execute-phase orchestrator or execute-plan command.
gsd-debugger
Investigates bugs using scientific method, manages debug sessions, handles checkpoints. Spawned by /gsd:debug orchestrator.
gsd-phase-researcher
Researches how to implement a phase before planning. Produces RESEARCH.md consumed by gsd-planner. Spawned by /gsd:plan-phase orchestrator.