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 Eliyce/paqad-ai --skill expert-need-detectorgit clone --depth 1 https://github.com/Eliyce/paqad-aiWrote 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/eliyce/paqad-ai/expert-need-detector)<a href="https://agentmods.dev/skills/eliyce/paqad-ai/expert-need-detector"><img src="https://agentmods.dev/badge/skills/eliyce/paqad-ai/expert-need-detector/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/eliyce/paqad-ai/expert-need-detector"><img src="https://agentmods.dev/badge/skills/eliyce/paqad-ai/expert-need-detector.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00087 | $0.01182 |
| Opus 5 | $0.00044 | $0.00591 |
| Sonnet 5 | $0.00017 | $0.00236 |
| Haiku 4.5 | $0.00009 | $0.00118 |
Grade A, and why
expert-need-detector 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 6d 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
What It Does
Reads the request and the pipeline's S0 grounding slice and decides which domain experts the spec needs — a database expert when the request touches the data model, a security expert when it touches auth or a trust boundary, a UI expert when it touches a screen, and so on. It returns a small JSON artifact naming each needed expert and, in one plain sentence, why it fired.
This is the model call that replaces a deterministic signal-scorer (issue #521, the "one change"): a script cannot reliably tell which expert a request needs — file-path and keyword heuristics emit false signals both ways — so the judgement is the model's. The script's job is only to VALIDATE the result against the roster, never to make it. Nothing needed ⇒ an empty list ⇒ zero experts, zero cost.
Use This When
- The spec pipeline is running with the expert roster enabled (
spec_pipeline_experts_enabled), after S0 grounding has producedgrounding.jsonand before the craft step.
Do not run this when the experts flag is off — with it off the pipeline is byte-identical to v1 and this skill never runs.
Inputs
request_text— required. The request being specced.grounding— required. The S0grounding.json(references + business terms) — the evidence for which areas the request touches. Decide from THIS, not from the whole repo.roster— optional. The allowed expert roles; defaults to the framework roster. You may name only roles in it. The roster and each role's remit are inruntime/base/skills/expert-need-detector/references/roster.md.
Procedure
- Read the request and the grounding terms/references. Identify the concrete areas the request touches (a table or migration, an auth path, a screen, an integration, an infra change).
- For each area that clearly needs a specialist, select the matching expert role from the roster only. Judge need, not certainty-of-self: pick an expert because the work plainly sits in its domain, never "to be safe".
- Select nothing when nothing clearly needs a specialist. An empty result is the common, correct outcome — it costs nothing downstream.
- For each selected expert write one plain-language
reasonnaming the area that triggered it (e.g. "adds the invoices migration", not "database concerns"). - Emit the JSON artifact (see Output Contract) and hand it to the pipeline:
paqad-ai spec pipeline experts record <artifact-file>. That command runs the deterministic roster guard (src/spec-pipeline/experts/need.ts) and refuses anything naming a role outside the roster — do not re-implement that check here.
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.
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.
- 6d ago First seen · 94 lines · 87 tokens per session scan A 15e5b2a0883e
expert-need-detector is a skill published in the GitHub repository Eliyce/paqad-ai (8 stars, last pushed yesterday), licensed MIT. It adds 87 tokens to every session and 1,182 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 2026-09-06.
Other skills, from other repositories
pr-writing-review
Extract and analyze writing improvements from GitHub PR review comments. Use when asked to show review feedback, style changes, or editorial improvements from a GitHub pull request URL. Handles both explicit suggestions and plain text feedback. Produces structured output comparing original phrasing with reviewer…
session-investigator
Investigate fast-agent session and history files to diagnose issues. Use when a session ended unexpectedly, when debugging tool loops, when correlating sub-agent traces with main sessions, or when analyzing conversation flow and timing. Covers session.json metadata, history JSON format, message structure, tool…
auto-go
A command that implements code from a SPEC, a document describing the required behavior and work.
auto-plan
A code-planning skill that examines a codebase and creates a detailed specification, implementation plan, and acceptance criteria. It can organize requirements using EARS, a structured way to describe how software should behave in different situations.
agent-pipeline
Multi-agent pipeline orchestration skill.
adaptive-quality
Per-task execution profile selection based on complexity in Balanced quality mode.