subagent-directives

subagent-directives is a skill for Claude Code, Codex from griddynamics/rosetta. It costs 13 tokens per session (314 once invoked), scanned A, original, Apache-2.0.

A set of mandatory instructions for a coding-agent subagent, meaning a smaller agent assigned part of a larger task.

In plain words
What is it for?
Use it to guide subagents through assigned tasks, required skills, evidence collection, checklists, and escalation to the orchestrator.
Why use it?
It keeps delegated work within scope, makes blockers visible, and requires evidence and honest reporting before the subagent finishes.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/griddynamics/rosetta/subagent-directives
Any agent
npx skills add griddynamics/rosetta --skill subagent-directives
Clone the repo
git clone --depth 1 https://github.com/griddynamics/rosetta

Made for: Claude Code, Codex.

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 subagent-directives

README.md
[![agentmods](https://agentmods.dev/badge/skills/griddynamics/rosetta/subagent-directives.svg)](https://agentmods.dev/skills/griddynamics/rosetta/subagent-directives)
Your own site
<a href="https://agentmods.dev/skills/griddynamics/rosetta/subagent-directives"><img src="https://agentmods.dev/badge/skills/griddynamics/rosetta/subagent-directives.svg" alt="Measured on agentmods" height="20"></a>
Per session 13 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 314 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00013 $0.00314
Opus 5 $0.00006 $0.00157
Sonnet 5 $0.00003 $0.00063
Haiku 4.5 $0.00001 $0.00031

Measured 2d ago against content hash e76467dee3b3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

subagent-directives 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.

instructions/r3/core/skills/subagent-directives/SKILL.md · 34 lines

What it actually says

<subagent_directives>

  1. USE the MUST skills from your prompt before starting; consider the RECOMMEND ones.
  2. EXECUTION_CONTROLLER directly requested in your prompt → MUST APPLY SKILL FILE assets/s-session-execution-controller.md.
  3. Look around to better understand the input — never deviate from the assigned task.
  4. Ambiguous instructions → STOP and ask orchestrator before executing.
  5. On any blocking condition — cannot execute as specified · off-plan · would exceed scope · other blocker — MUST STOP + explain + report to orchestrator; never improvise beyond scope.
  6. Close out against the prompt's Checklist; report honestly: deviations, assumptions, open items (coded ≠ done).
  7. Return EXACTLY per Output specs — nothing missing, nothing extra.
  8. Provide proofs per Evidence specs: claims → deep links + line ranges + brief quotes; facts ≠ assumptions.
  9. Subagents ask orchestrator; orchestrator asks user.
  10. MUST split work into groups for factory-style tasks (example: cover unit test => read one small set of files - build tests, only THEN read another small file).
  • Silently continuing when blocked.
  • Assuming context not provided in prompt.

</subagent_directives>

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 2d ago First seen · 34 lines · 13 tokens per session scan A e76467dee3b3

Subscribe to this mod's changes

subagent-directives is a skill published in the GitHub repository griddynamics/rosetta (342 stars, last pushed yesterday), licensed Apache-2.0. It adds 13 tokens to every session and 314 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 2026-09-03.