skill-accuracy

A documentation rule for checking that descriptions of skills match their actual code and configuration.

In plain words
What is it for?
Use it when documenting skill categories, security claims, or language-model classification, including the model, fallback, threshold, and failure behavior.
Why use it?
It reduces errors such as documenting unsupported types, incorrect encryption, or incomplete details about how an AI classifier behaves.

Cursor rule for Cursor

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.

agentmods
npx agentmods add rules/wyattowalsh/agents/skill-accuracy
Clone the repo
git clone --depth 1 https://github.com/wyattowalsh/agents

Made for: Cursor.

Per session 9 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 202 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00009 $0.00202
Opus 5 $0.00005 $0.00101
Sonnet 5 $0.00002 $0.00040
Haiku 4.5 $0.00001 $0.00020

Measured 2d ago against content hash 41a2b5a69322, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

skill-accuracy 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 2d 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.

.cursor/rules/skill-accuracy.mdc · 19 lines

What it actually says

When documenting skills, verify accuracy against actual implementation:

Type Inventory: Read classifier prompts, keyword fallbacks, and DB enum definitions before documenting enum values or category lists. Types that exist only in documentation but not in code are phantom types — they cause silent failures.

Encryption/Security Claims: Verify against actual cryptography library calls (e.g., "AES-256" claimed but Fernet/AES-128-CBC used). Read imports, not comments.

LLM Classifier Documentation: Document all four of:

  1. Model — exact model identifier
  2. Fallback mode — behavior when API key missing or call fails
  3. Confidence threshold — minimum score to act on a result
  4. Failure handling — how to handle classification failures
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. 2d ago First seen · 19 lines · 202 tokens per session scan A 41a2b5a69322

Subscribe to this mod's changes

skill-accuracy is a cursor rule published in the GitHub repository wyattowalsh/agents (5 stars, last pushed 11d ago), licensed MIT. It adds 9 tokens to every session and 202 once invoked, about $0.0000 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.