Show HN: Decide – decision model CLI for analyzing large codebases with Jev Deepnoodle-ai released decide v0.5.0, an open-source decision-model CLI that splits codebases, diffs, JSONL, CSV, Markdown and folders of images or PDFs into individual items and asks a decision model a yes/no question about each, caching answers so repeat runs only query what changed. The tool ships 12 built-in question templates, splits Go, Python, JavaScript, TypeScript and Java into functions with their imports, and can exit with code 2 on flagged results or emit JSON, CSV, Markdown or GitHub annotations; it is installable via Homebrew, go install or go get, and as a Claude Code plugin with /decide:audit and /decide:hunt skills. Deepnoodle-ai cites $0.042 per million input tokens on the Jev decision model, which bills input only, versus $0.10 per million input tokens for gemini-decision-1, arguing decision models return one calibrated answer without generating output tokens. v0.5.0 OpenAI's Decisions API and Go templates → Judgment calls, at scale. Find the vulnerable function in a codebase, the urgent ticket in a queue, the risky change in a pull request. decide asks a decision model about every one, from your shell, CI, Go code or Claude Code. bash $ brew install deepnoodle-ai/tap/decide bash $ go install github.com/deepnoodle-ai/decide/cmd/decide@latest bash $ go get github.com/deepnoodle-ai/decide /plugin marketplace add deepnoodle-ai/decide /plugin install decide@decide The small calls inside your automations. Every pipeline has judgment calls that are too many for a person and too fuzzy for a rule. Here are a few that decide makes out of the box. What you’d otherwise build yourself. A decision model’s API takes one text and one question. decide brings the questions, splits your data into items, and gives your script something to act on. Questions, already written Twelve built-in templates, each a set of questions with tuned wording and flag thresholds. Write your own in a few lines of JSON. $ decide templates Code, function by function decide splits Go, Python, JavaScript, TypeScript and Java into functions and sends each with its imports. On a diff, it asks only about the functions that changed. $ decide run code-risk src --each function Any data, item by item Point it at JSONL, CSV, Markdown, a diff, or a folder of images or PDFs. Each record, section, hunk or file gets its own answer, and long ones are judged in parts. $ decide run relevance docs --each section No question asked twice Every answer is kept. Run again and decide asks only about what changed, so a second pass over a codebase costs almost nothing. 81974 answers from cache · 450 asked Output a script can act on A flagged answer can exit 2. Results print as JSON, CSV, a Markdown report or GitHub annotations, and a stopped run resumes where it left off. --fail-on flagged --format github The same from Go Typed questions from your own program, with validated answers, retries, a fake server for tests, and patterns such as gate, rank and funnel. $ go get github.com/deepnoodle-ai/decide Audit a whole codebase from Claude Code. Two skills in the decide plugin. decide scores every function, and Claude reads only the few that score highest. - /decide:audit Audits a repository for security bugsdecide scores every function with the security template. Claude reads up to 15 of the highest with their callers, and reports which are real. - /decide:hunt Finds where a fixed bug was repeatedGive it a fix commit or pull request. Claude writes one question about the mistake, decide asks it of every function, and Claude reads the top matches. In Claude Code /plugin marketplace add deepnoodle-ai/decide /plugin install decide@decide Needs the decide CLI and a key for a decision model. Once installed, the plugin also checks what Claude reads and runs. Claude Code guide → https://decide.deepnoodle.ai/reference/claude-code/ Why not just call an LLM? You can, and for writing or reasoning you should. For a yes or a no on every item, a decision model is the better tool. Faster An LLM writes its reply a token at a time, and you wait for all of it. A decision model writes nothing. It returns one answer. Cheaper With an LLM you pay for your prompt and for every token it writes back. Here you pay for the prompt. $0.042 per million input tokens on Jev 1 price-note Calibrated Ask an LLM how sure it is and it types a number, usually a confident one. A decision model is trained so its probability matches how often it is right. 90% means right about nine times in ten