request-classifier

request-classifier is a skill for Codex from Eliyce/paqad-ai. It costs 14 tokens per session (570 once invoked), scanned A, original, MIT.

A request-routing step that classifies new work by its workflow, affected area, size, and risk. It uses fixed allowed categories so later skills and checks can be selected consistently.

In plain words
What is it for?
Use it when a request starts or changes enough that earlier assumptions about its scope may no longer apply.
Why use it?
It reduces guesswork about how a request should be handled and favors cautious decisions when the scope or risk is unclear.

Skill for Codex

Written for Codex: agents/openai.yaml present.

Good fit Use it when a request starts or changes enough that earlier assumptions…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/eliyce/paqad-ai/request-classifier
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.

Any agent
npx skills add Eliyce/paqad-ai --skill request-classifier
Clone the repo
git clone --depth 1 https://github.com/Eliyce/paqad-ai

Made for: 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 request-classifier

README.md
[![agentmods](https://agentmods.dev/badge/skills/eliyce/paqad-ai/request-classifier.svg)](https://agentmods.dev/skills/eliyce/paqad-ai/request-classifier)
Your own site
<a href="https://agentmods.dev/skills/eliyce/paqad-ai/request-classifier"><img src="https://agentmods.dev/badge/skills/eliyce/paqad-ai/request-classifier.svg" alt="Measured on agentmods" height="20"></a>
Per session 14 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 570 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00014 $0.00570
Opus 5 $0.00007 $0.00285
Sonnet 5 $0.00003 $0.00114
Haiku 4.5 $0.00001 $0.00057

Measured 7d ago against content hash a9e70c760f70, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

request-classifier 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 7d ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/extract-signals.sh, scripts/lint-output.sh), 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.

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.

runtime/base/skills/request-classifier/SKILL.md · 74 lines

What it actually says

What It Does

Produces a deterministic request classification across the routing dimensions used by the framework so downstream skill and rule selection can rely on explicit signals instead of guesswork.

Use This When

Use this at the start of a new request or whenever the effective scope changed enough that prior routing assumptions may no longer be safe.

Inputs

  • Read the request text, attached plan, and any directly linked issue or spec.
  • Read the nearest canonical docs only when they clarify stack, module, or risk signals.
  • Read references/decision-rules.md before assigning ambiguous dimension values.

Procedure

  1. Pipe the raw request into scripts/extract-signals.sh to get baseline keyword-derived dimensions (workflow, ui/api/db impact, scope, risk hint).
  2. Override script defaults only when canonical evidence (linked spec, stack manifest, module map) contradicts them — record the override under Evidence.
  3. Use assets/dimensions.txt for the allowed value set per dimension.
  4. Default to most conservative accurate value when ambiguous; never invent certainty for risk/sensitivity.
  5. Format per assets/output.template.md; validate with scripts/lint-output.sh.

Output Contract

  • Match assets/output.template.md: ## Classification (one key: value per dimension) and ## Evidence (bullet list for non-obvious calls).
  • Output must pass scripts/lint-output.sh (exit 0).

Escalate / Stop Conditions

  • Ask when stack, workflow, or impact dimensions cannot be inferred safely from the request.
  • Warn when canonical docs contradict the request about modules, interfaces, or current behavior.
  • Do not fabricate certainty for risk or sensitivity dimensions.

Resources

  • references/decision-rules.md
  • scripts/extract-signals.sh
  • scripts/lint-output.sh
  • assets/output.template.md
  • assets/dimensions.txt
  • runtime/base/skills/existing-doc-checker/SKILL.md
  • agents/openai.yaml
Files

What ships with it

6 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. 7d ago First seen · 74 lines · 14 tokens per session scan A a9e70c760f70

Subscribe to this mod's changes

request-classifier is a skill published in the GitHub repository Eliyce/paqad-ai (8 stars, last pushed today), licensed MIT. It adds 14 tokens to every session and 570 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 2026-08-31.