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 skills add mifunedev/openharness --skill delegategit clone --depth 1 https://github.com/mifunedev/openharnessWrote 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/skills/mifunedev/openharness/delegate)<a href="https://agentmods.dev/skills/mifunedev/openharness/delegate"><img src="https://agentmods.dev/badge/skills/mifunedev/openharness/delegate/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/skills/mifunedev/openharness/delegate"><img src="https://agentmods.dev/badge/skills/mifunedev/openharness/delegate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00081 | $0.03696 |
| Opus 5.5 | $0.00032 | $0.01478 |
| Sonnet 5.5 | $0.00016 | $0.00739 |
| Haiku 4.5 | $0.00008 | $0.00370 |
Grade A, and why
delegate 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 1mo 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 — 323 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delegate
Parallel execution coordinator. Read a plan or conversation context, decompose it into a dependency-ordered task graph, and spawn worker sub-agents in parallel waves. Each wave completes before the next begins. Results are collected, validated, and reported.
Core principle: maximize parallelism while respecting dependencies absolutely.
When a worker is justified
Use a worker only when the task is self-contained and gains from parallelism, isolated context, restricted tools, or containment of verbose disposable output. Keep work in the active session when phases share substantial context or require iterative refinement.
A worker is a bounded execution context, not a project role. The active coding
agent is the runtime and stays the owner of the work; skills — /architect,
/spec, /audit, /retro — are how it adopts a role. Delegation buys isolation
or parallelism and nothing else.
| Delegate it | Keep it in the active session |
|---|---|
| Independent parallel research or source sweeps | Planning, implementation, and testing that share one evolving picture |
| Verbose disposable output — logs, search dumps, test runs | Work that needs iterative refinement against operator feedback |
| Disjoint file ownership with no shared mutable state | Two tasks that touch the same file |
| A deliberate tool or permission restriction | Anything whose result you would have to re-derive to use |
Do not invent named architectural roles for workers. /delegate owns fan-out
policy; other skills must not grow a competing worker hierarchy beside it.
Worker model and thinking policy
Apply this policy to every worker:
- Inherit the parent/session model by default. Omit the Agent tool's
modelargument. Do not route routine or simple work to a weaker model tier. - Set the Agent tool's
thinkingparameter from task complexity: simple/mechanical →low, standard →medium, complex →high, and architecture or debugging with substantial uncertainty →xhigh. Supported levels areoff,minimal,low,medium,high, andxhigh; never usemax. - If the selected thinking level is unsupported by the inherited model/provider, use the nearest supported level. Do not switch models merely to obtain a thinking level.
- Override
modelonly with an explicit task-specific reason: an operator request, an unavailable required capability/context, a strict latency or budget constraint, or local benchmark evidence. Record that reason in the task graph and pass the override only for that worker.
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.
- 1mo ago Changed · +57 lines · +14 tokens per session fb9f5f399564
- 1mo ago First seen · 266 lines · 67 tokens per session scan A a6c952464278
delegate is a skill published in the GitHub repository mifunedev/openharness (37 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 81 tokens to every session and 3,696 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
data-visualization
Use for creating publication-quality charts and multi-panel analysis summaries. Triggers when tasks involve visualizing data, plotting results, creating charts, or producing visual reports from analysis output.
cuml-machine-learning
Use for GPU-accelerated machine learning on tabular data using NVIDIA cuML. Triggers when tasks involve classification, regression, clustering, dimensionality reduction, or model training on datasets.
blog-post
Writes and structures long-form blog posts, creates tutorial outlines, and optimizes content for SEO with cover image generation. Use when the user asks to write a blog post, article, how-to guide, tutorial, technical writeup, thought leadership piece, or long-form content.
social-media
Drafts engaging social media posts, writes hooks, suggests hashtags, creates thread structures, and generates companion images. Use when the user asks to write a LinkedIn post, tweet, Twitter/X thread, social media caption, social post, or repurpose content for social platforms.
schema-exploration
Lists tables, describes columns and data types, identifies foreign key relationships, and maps entity relationships in a database. Use when the user asks about database schema, table structure, column types, what tables exist, ERD, foreign keys, or how entities relate.
deepagents-thread-inspector
Inspect and explain conversations in the local Deep Agents Code SQLite session store. Use to look up the current thread's title or name, as a fallback when LangSmith trace tooling is unavailable, for offline or untraced sessions, or when asked to identify or summarize a local dcode thread, inspect checkpoint metadata…