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 skills/deephaven/deephaven-mcp/ref-skill-effectivenessnpx skills add deephaven/deephaven-mcp --skill ref-skill-effectivenessgit clone --depth 1 https://github.com/deephaven/deephaven-mcpWrote 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/deephaven/deephaven-mcp/ref-skill-effectiveness)<a href="https://agentmods.dev/skills/deephaven/deephaven-mcp/ref-skill-effectiveness"><img src="https://agentmods.dev/badge/skills/deephaven/deephaven-mcp/ref-skill-effectiveness.svg" alt="Measured on agentmods" 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 | $0.00073 | $0.03170 |
| Opus 5 | $0.00036 | $0.01585 |
| Sonnet 5 | $0.00015 | $0.00634 |
| Haiku 4.5 | $0.00007 | $0.00317 |
Grade A, and why
ref-skill-effectiveness 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 3d 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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Effectiveness
A skill that the agent never invokes is dead code. A skill the agent invokes but cannot act on is worse. This document defines the content properties that make agent-facing prose actually drive behavior. It is loaded by ref-skill-authoring-standards, ref-agents-md-curation, skill-add, and skill-review.
Effectiveness is the goal; structure is in service of effectiveness. A perfectly-structured skill the agent never invokes is worse than a structurally-imperfect skill that fires correctly every time. The published sources and their precedence are in ref-skill-authoring-standards Sources; rules here cite them where they derive from them.
Triggerability
The frontmatter description is the only text the agent sees at decision time. If it does not name the moment, the skill is dark code.
- Required content in every description:
- Third person, per the spec — the description is catalog metadata about the skill, not an instruction to the agent.
- What the skill does, stated plainly.
- When to use it — the trigger: an artifact, a verb-noun task, a named workflow, a moment in a process.
- What the skill is not for when ambiguity with a sibling skill exists. When a skill already carries a "When to use this vs. X" line in its body, that disambiguator belongs in the description too — routing happens at Level 1, and a body-only disambiguator arrives after the routing decision is already made.
- Front-load the trigger. Hosts shorten long descriptions when the catalog is large, and the trigger clause is the part that must survive. A description whose "when" sits in the final clause is one truncation away from being a topic label.
- Triggerability test: read the skill's description alongside its ten nearest sibling descriptions in the catalog README. Given a representative task, can you predict — without reading any body — which skill should fire? If no, the description is broken.
- Acceptable / unacceptable pairs:
- ✅ "Add a new command to the dhcli CLI — invoke when adding or editing a verb under cli/_commands/. Wraps the click + Pattern B + structured-error + agents-manifest conventions; prevents the most common bugs." (Names the verb, the artifact, the triggering moment, and the failure mode it prevents — with the trigger ahead of the detail.)
- ❌ "CLI command guidance." (Names the topic; not the moment.)
- ✅ "Run a single test file's tests with coverage — required for assessing per-file coverage of a single source file." (Names the verb, the artifact, and the disambiguator from
tests-run.) - ❌ "Test running helper." (No verb, no disambiguator, no artifact.)
- ✅ "Verify a markdown documentation file is factually accurate — invoke for surgical correctness-only fixes when the document's structure is already sound; use docs-improve for a full review. Checks commands, file paths, config keys, API names, code examples, and URLs against source code, and fixes inaccuracies in place." (Names verb, artifact, trigger, the sibling disambiguator, scope, and side effect.)
- ❌ "Documentation review skill." (All three failures.)
- Empirical grounding: the longest-standing effectiveness bug in this codebase was
ErrorCode.help_textemitting one identical string per member because the description-level__doc__was read as a member-level attribute — a triggerability failure at the catalog layer (dhcli agents errorsreported "10 codes, 1 unique help string"). The skill-level analog is a description that names the topic but not the moment.
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.
- 3d ago First seen · 111 lines · 73 tokens per session scan A 0aa6e5f09f2a
ref-skill-effectiveness is a skill published in the GitHub repository deephaven/deephaven-mcp (5 stars, last pushed 6d ago), licensed Apache-2.0. It adds 73 tokens to every session and 3,170 once invoked, about $0.0004 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…