agentmods.dev

sanad-agent: Skill for Claude Code

.agents/skills/delegate-task-supervisor/SKILL.md

delegate-task-supervisor is a skill for Claude Code, Codex from EastStarAI/sanad-agent. It costs 116 tokens per session (6,543 once invoked), scanned C, original, MIT.

A coordinator for handing coding tasks to external agent tools such as OpenCode and Antigravity. It keeps track of workspaces, sessions, parallel tasks and task progress while the delegated tools handle their own implementation instructions.

In plain words
What is it for?
Use it to dispatch one or more coding tasks, maintain isolated work areas for parallel work and observe delegated task timelines.
Why use it?
It prevents delegated tasks from losing project or session identity and avoids conflicting work in the same conversation. It also provides a way to wait for meaningful task updates without repeatedly checking manually.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions subagents; installed under .agents/ (shared by several agents); mentions OpenCode.

This is EastStarAI/sanad-agent's own configuration. It tells Claude Code and Codex how to work on sanad-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 sanad-agent configures →

not rated 47repo +1 changed 14d ago C scan 116 tokens original MIT
View source ↗ EastStarAI/sanad-agent sanad.eaststarai.com
Reuse

Borrowing it

Nothing to install: this file belongs to EastStarAI/sanad-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/EastStarAI/sanad-agent/main/.agents/skills/delegate-task-supervisor/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/EastStarAI/sanad-agent

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 delegate-task-supervisor

README.md
[![agentmods](https://agentmods.dev/badge/skills/eaststarai/sanad-agent/delegate-task-supervisor/github.svg)](https://agentmods.dev/skills/eaststarai/sanad-agent/delegate-task-supervisor)
Your own site
<a href="https://agentmods.dev/skills/eaststarai/sanad-agent/delegate-task-supervisor"><img src="https://agentmods.dev/badge/skills/eaststarai/sanad-agent/delegate-task-supervisor/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 delegate-task-supervisor

Your own site · 80×15
<a href="https://agentmods.dev/skills/eaststarai/sanad-agent/delegate-task-supervisor"><img src="https://agentmods.dev/badge/skills/eaststarai/sanad-agent/delegate-task-supervisor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 116 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 6,543 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 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 $0.00116 $0.06543
Opus 5.5 $0.00046 $0.02617
Sonnet 5.5 $0.00023 $0.01309
Haiku 4.5 $0.00012 $0.00654

Measured 14d ago against content hash c8b2398bd694, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade C, and why

delegate-task-supervisor 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 14d ago.

The scan reads SKILL.md. This mod also ships 28 executable files (scripts/bootstrap.mjs, scripts/supervisor-control.mjs, scripts/supervisor-core.mjs, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Tells the agent never to refusehighAnti-refusal

Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.

- **`turn_in_flight` (exit code 3):** The target session is still owned by an active durable turn (`running`, `resuming`, `waiting`, or `stopping`). The supervisor must not immediately re-dispatch the same task or send a
.agents/skills/delegate-task-supervisor/SKILL.md · 318 lines

How it starts

The opening of the file, as written. The whole thing — 318 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Delegate Task Supervisor

Coordinate external implementers without duplicating their own delegation contracts. This skill owns bootstrap, workspace identity, concurrency, durable task state, human observability, dynamic task queuing, and event-driven wake-up. The selected *-delegate skill continues to own briefs, permissions, relay flags, result semantics, review, and resumption.

Boundaries

  • Supported implementers are opencode, agy/antigravity, and sanad.
  • Load and follow agy-delegate for Antigravity, opencode-delegate for OpenCode, and sanad-delegate for Sanad Agent. Do not copy or reinterpret their operational instructions.
  • Use delegate-setup only when the user wants reusable fleet lanes. Direct one-off dispatch does not require a lane.
  • For Sanad implementation, also load Sanad Subagent Developer; use an isolated worktree for every parallel write task.
  • Never run two tasks concurrently in the same OpenCode session, Antigravity conversation, or Sanad session.
  • Never commit, push, merge, or open a pull request unless the user explicitly authorizes that delivery action.

<skill-dir> below means the directory containing this file.

1. Bootstrap once per machine

Check prerequisites and installed integrations:

node "<skill-dir>/scripts/bootstrap.mjs" check

check distinguishes the installed Sanad CLI from a validated source-development fallback. When the installed binary is absent but the invoking checkout contains agent/bin/sanad_agent.dart and fvm (or dart) resolves on PATH, the report names the fallback as validated and actionable instead of treating the missing CLI as a blocker; install-sanad is then only needed when a release binary is required. An installed sanad command always remains the preferred delegation path.

If anything is missing, show the complete planned changes and obtain explicit approval. Then run:

node "<skill-dir>/scripts/bootstrap.mjs" install --yes

The installer adds only missing dependencies. Authentication remains user-owned: after installation, complete any browser/keyring login requested by the implementer CLI, then rerun check. When sanad is missing, refer users to the install-sanad skill; the bootstrap script does not attempt automatic Sanad installation.

Read the full file on GitHub · 318 lines

Files

What ships with it

29 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. 14d ago Changed · +61 lines · +12 tokens per session scan A → C c8b2398bd694
  2. 15d ago Changed 525d02b11ee4
  3. 16d ago Changed · +30 lines · +8 tokens per session 10924fc9ddaa
  4. 1mo ago Changed · +5 lines 929e41387330
  5. 1mo ago Changed · +75 lines a5da900cfcfd
  6. 1mo ago First seen · 147 lines · 96 tokens per session scan A 239160119f5c

Subscribe to this mod's changes

delegate-task-supervisor is a skill published in the GitHub repository EastStarAI/sanad-agent (47 stars, last pushed yesterday), licensed MIT. It adds 116 tokens to every session and 6,543 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-22.

Building agentsai-agentcoding-agentdartdesktopdeveloper-toolsflutter
Related

delegate-task-supervisor compared with skills from other repositories

delegate-task-supervisor next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
delegate-task-supervisorEastStarAI/sanad-agentSkill47yesterdayMITC116Claude Code, Codex
flutterpledgeandgrow/pledge-skillsSkill103mo agoMITA126Claude Code, Codex
at_client_skills-sdkatsign-foundation/at_client_sdkSkill1.4kyesterdayBSD-3-ClauseA220Claude Code
flutter-pre-cachingevanca/flutter-ai-rulesSkill669yesterdayMITA36Claude Code
firebase-authevanca/flutter-ai-rulesSkill669yesterdayMITA29Claude Code
firebase-messagingevanca/flutter-ai-rulesSkill669yesterdayMITA35Claude Code
code-reviewevanca/flutter-ai-rulesSkill669yesterdayMITA27Claude Code