cjk-structured-llm-output-truncates-json-needs-2x-tokens SKILL.md skillA
cjk-structured-llm-output-truncates-json-needs-2x-tokens from wan-huiyan/agent-traffic-control is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add wan-huiyan/agent-traffic-control/cjk-structured-llm-output-truncates-json-needs-2x-tokens@v1.0.0Pins v1.0.0 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- wan-huiyan/agent-traffic-control
- Path
plugins/agent-traffic-control/skills/cjk-structured-llm-output-truncates-json-needs-2x-tokens/SKILL.md- Versions recorded
- 1
- Latest version
- v1.0.0, committed Jul 8, 2026
- Size
- 5,154 bytes
- Repository stars
- 3
- License
- MIT
When generating LONG Chinese/Japanese/Korean (CJK) STRUCTURED output (a big JSON report, a multi-section document) from an LLM chat API, the response truncates mid-JSON and your parser throws "no parseable JSON" / "Expecting ',' delimiter" / JSONDecodeError — even though the SAME prompt in English worked. Root cause: CJK is token-heavy (≈1–2 tokens per character vs ≈0.25 for English), so a report that fit your max_tokens in English overflows it in Chinese. Some models ALSO hard-cap output well…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Long CJK structured output truncates JSON — budget ~2× tokens, and salvage the rest
## Problem
You ask an OpenAI-compatible LLM for a long JSON report. In English it returns valid JSON. In **Chinese
(or Japanese/Korean)** the identical request returns a string that **starts** with valid JSON but is **cut off
mid-object** → `json.loads` fails, your regex `\{.*\}` grabs an unbalanced blob, and you get "no parseable JSON".
## Why
- **CJK is token-dense.** One Han character is often 1–2 BPE tokens; English averages ~0.25 tokens/char. The
*same* report is ~1.5–2× more output tokens in Chinese. A `max_tokens` that comfortably fit the English report
truncates the Chinese one.
- **Per-model output caps differ and some are low.** Requesting `max_tokens` above a model's hard cap is either
rejected (`400 ... Range of max_tokens should be [1, 8192]`) or silently clamped, then the long CJK report
truncates. Observed: **qwen-max hard-caps at 8192** (can't emit a full Chinese report); glm-4.6 / minimax ≈16K;
qwen-plus / qwen3-235b / glm-5.2 / kimi / deepseek handle 24–32K. Reasoning models also spend tokens on
`reasoning_content` *before* the JSON, compounding it.
## Fix
…Read the whole file at its exact version.
How to install
mdr add wan-huiyan/agent-traffic-control/cjk-structured-llm-output-truncates-json-needs-2x-tokens@sha256:d7218f2a8f2c1b9cThe line at the top of this page pins the label v1.0.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_gkravpaesui24sdf)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Jul 8, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (5154 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
wan-huiyan/agent-traffic-control · 3 stars · license MIT · pushed 2026-09-22 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_gkravpaesui24sdf GET https://markdownregistry.com/api/v1/resolve?ref=wan-huiyan/agent-traffic-control/cjk-structured-llm-output-truncates-json-needs-2x-tokens GET https://markdownregistry.com/api/v1/blob/d7218f2a8f2c1b9c544beaf4617610a7787224edbc3445f2726402efc515a053
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.