roli-lpci/lintlang
Static linter for AI agent tool descriptions, system prompts, and configs. Catches vague instructions, missing constraints, and schema mismatches before they cause agent misbehavior.
Implements seven structural detectors (H1-H7) that identify language-specific failure patterns—tool ambiguity, missing constraints, schema mismatches, context leakage, implicit instruction failures, template violations, and role confusion—each mapped to real-world agent breakdowns. Works as a pure static analyzer with zero LLM calls, outputting verdict-based verdicts (PASS/REVIEW/FAIL) with concrete fix suggestions rather than scores. Integrates directly into CI pipelines via CLI with severity filtering and supports YAML, JSON, and plain-text prompt files across major agent frameworks (OpenAI, Anthropic, OpenHands).
Available on PyPI.
Stars
4
Forks
1
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 02, 2026
Monthly downloads
265
Commits (30d)
0
Dependencies
1
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/roli-lpci/lintlang"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
connectaman/LoPace
LoPace is a bi-directional encoding framework designed to reduce the storage footprint of...
LakshmiN5/promptqc
ESLint for your system prompts — catch contradictions, anti-patterns, injection vulnerabilities,...
sbsaga/toon
TOON — Laravel AI package for compact, human-readable, token-efficient data format with JSON ⇄...
nooscraft/tokuin
CLI tool – estimates LLM tokens/costs and runs provider-aware load tests for OpenAI, Anthropic,...
thesupermegabuff/megabuff-cli
🤖 CLI for Better prompts, transparent costs, & zero vendor lock-in. Optimize AI prompts across...