# hunch - Doramagic AI Context Pack > Positioning: a pre-install experience and judgment asset. It helps the host AI get off to a good start, but it does not mean the project has already been installed, run, or validated. ## Sufficiency Principle - **Sufficiency over compression**: The AI Context Pack should be sufficient for the host AI to understand the project's value, capability boundaries, entrypoints, risks, and evidence sources before starting work; it may be layered, but it does not aim for the shortest possible summary. - **Compression policy**: Compress only noise and duplication, never context that affects judgment or the quality of the work. ## How the Host AI Should Use This You are reading the AI Context Pack that Doramagic compiled for hunch. Treat it as pre-work context: help the user understand who it fits, what it can do, how to start, what must be verified after install, and where the risks are. Do not claim that you have already installed, run, or executed the target project. ## Claim Consumption Rules - **Fact source**: Repo Evidence + Claim/Evidence Graph; the Human Wiki only supplies salience, terminology, and narrative structure. - **Minimum status for a fact**: `supported` - `supported`: May be used as a project fact, but the answer must cite the claim_id and evidence path. - `weak`: Usable only as a low-confidence lead; the user must be asked to keep verifying. - `inferred`: Usable only for risk notes or open questions; must not be packaged as a project fact. - `unverified`: Must not be used as fact; state clearly that evidence is insufficient. - `contradicted`: Must show the conflicting sources and must not force a single version on the user's behalf. ## Who It Fits Best - **Developers already using host AIs such as Claude/Codex/Cursor/Gemini**: The README or plugin config mentions multiple host AIs. Evidence: `README.md` Claim: `clm_0004` supported 0.86 - **Users who want to bring professional workflows into a host AI**: The repo contains Skill documents. Evidence: `.agents/skills/fable-mode/SKILL.md`, `.agents/skills/source-command-hunch-fragile/SKILL.md`, `.claude/skills/fable-mode/SKILL.md`, `plugin/skills/capture/SKILL.md` et al. Claim: `clm_0005` supported 0.86 ## What It Can Do - **AI Skill / Agent Instruction Asset Library** (Previewable before install): The project contains Skill or Agent instruction files that a host AI can read, useful for bringing professional workflows into hosts like Claude, Codex, or Cursor. Evidence: `.agents/skills/fable-mode/SKILL.md`, `.agents/skills/source-command-hunch-fragile/SKILL.md`, `.claude/skills/fable-mode/SKILL.md`, `plugin/skills/capture/SKILL.md` et al. Claim: `clm_0001` supported 0.86 - **Multi-Host Install and Distribution** (Verify after install): The project contains plugin or marketplace configuration, indicating it targets install and distribution across one or more AI hosts. Evidence: `.claude-plugin/marketplace.json`, `plugin/.claude-plugin/plugin.json` Claim: `clm_0002` supported 0.86 - **Command-Line Startup or Install Flow** (Verify after install): The project documentation contains runnable commands; real use requires running them in a local or host environment. Evidence: `README.md` Claim: `clm_0003` supported 0.86 ## How to Start - `npm i -g @davesheffer/hunch` Evidence: `README.md` Claim: `clm_0006` supported 0.86 ## Continue-or-Stop Decision Card - **Current recommendation**: Sandbox-trial permissions first - **Why**: The project has signals of install commands, host configuration, or local writes; do not go straight into your primary environment—trial it in isolation first. ### 30-Second Read - **What to do now**: Sandbox-trial permissions first - **Minimum safe next step**: Run Prompt Preview first; if you still want to install, trial only in an isolated environment - **Do not trust yet**: Tool permission boundaries cannot be trusted before install. - **Continuing will touch**: Command execution, Host AI configuration, Local environment or project files ### What You Can Trust Now - **Target-audience signal: Developers already using host AIs such as Claude/Codex/Cursor/Gemini** (supported): Backed by a supported claim or project evidence, but that still is not the same as real install results. Evidence: `README.md` Claim: `clm_0004` supported 0.86 - **Target-audience signal: Users who want to bring professional workflows into a host AI** (supported): Backed by a supported claim or project evidence, but that still is not the same as real install results. Evidence: `.agents/skills/fable-mode/SKILL.md`, `.agents/skills/source-command-hunch-fragile/SKILL.md`, `.claude/skills/fable-mode/SKILL.md`, `plugin/skills/capture/SKILL.md` et al. Claim: `clm_0005` supported 0.86 - **Capability exists: AI Skill / Agent Instruction Asset Library** (supported): You can trust that the project contains signals of this capability; whether it fits your specific task still needs trial or after-install verification. Evidence: `.agents/skills/fable-mode/SKILL.md`, `.agents/skills/source-command-hunch-fragile/SKILL.md`, `.claude/skills/fable-mode/SKILL.md`, `plugin/skills/capture/SKILL.md` et al. Claim: `clm_0001` supported 0.86 - **Capability exists: Multi-Host Install and Distribution** (supported): You can trust that the project contains signals of this capability; whether it fits your specific task still needs trial or after-install verification. Evidence: `.claude-plugin/marketplace.json`, `plugin/.claude-plugin/plugin.json` Claim: `clm_0002` supported 0.86 - **Capability exists: Command-Line Startup or Install Flow** (supported): You can trust that the project contains signals of this capability; whether it fits your specific task still needs trial or after-install verification. Evidence: `README.md` Claim: `clm_0003` supported 0.86 - **There are Quick Start / install-command signals** (supported): You can trust that the docs mention a startup or install entrypoint; do not run it directly in your primary environment because of that. Evidence: `README.md` Claim: `clm_0006` supported 0.86 ### What You Cannot Trust Yet - **Tool permission boundaries cannot be trusted before install.** (unverified): MCP/tool projects usually touch files, the network, the browser, or external APIs, so permissions and logs must be checked for real. - **Real output quality cannot be trusted before install.** (unverified): Prompt Preview can only show how it guides you; it cannot prove result quality in the real project. - **Host AI version compatibility cannot be trusted before install.** (unverified): Host loading rules and version differences across Claude, Cursor, Codex, Gemini, and others must be verified in a real environment. - **That it will not pollute your existing host AI's behavior cannot be trusted directly.** (inferred): Skill, plugin, and AGENTS/CLAUDE/GEMINI instructions may change the host AI's default behavior. Evidence: `.agents/skills/fable-mode/SKILL.md`, `.agents/skills/source-command-hunch-fragile/SKILL.md`, `.claude-plugin/marketplace.json`, `.claude/skills/fable-mode/SKILL.md` et al. - **Safe rollback cannot be assumed by default.** (unverified): Unless the project clearly provides uninstall and recovery instructions, verify in an isolated environment first. - **After a real install, is it compatible with the user's current host AI version?** (unverified): Compatibility can only be verified in the actual host environment. Evidence: `.claude-plugin/marketplace.json`, `plugin/.claude-plugin/plugin.json` - **Does the project's output quality meet the user's specific task?** (unverified): The pre-install preview can only show flow and boundaries; it cannot replace real evaluation. - **Do the install commands require network access, permissions, or global writes?** (unverified): This affects install risk in both enterprise and personal environments. Evidence: `README.md` ### What Continuing Will Touch - **Command execution**: Package managers, network downloads, the local plugin directory, project config, or the user's home directory. Why: Running the very first command can already change your environment; decide whether it is worth running first. Evidence: `README.md` - **Host AI configuration**: The plugin, Skill, or rule-loading config of hosts like Claude/Codex/Cursor/Gemini/OpenCode. Why: Host configuration changes how the AI works afterward and may conflict with the user's existing rules. Evidence: `.agents/skills/fable-mode/SKILL.md`, `.agents/skills/source-command-hunch-fragile/SKILL.md`, `.claude-plugin/marketplace.json`, `.claude/skills/fable-mode/SKILL.md` et al. - **Local environment or project files**: Install results, plugin caches, project config, or local dependency directories. Why: The write scope and rollback path cannot be proven before install and need isolated verification. Evidence: `.claude-plugin/marketplace.json`, `README.md`, `plugin/.claude-plugin/plugin.json` - **Host AI context**: The AI Context Pack, Prompt Preview, Skill routing, risk rules, and project facts. Why: Importing context affects the host AI's later judgment, so avoid packaging unverified items as facts. ### Minimum Safe Next Steps - **Run Prompt Preview first**: Use a pre-install interactive trial to judge whether the way of working fits; it needs no authorization or environment change. (applies when: Applies to any project, especially when output quality is unknown.) - **Trial-install only in an isolated directory or a test account**: Avoid letting install commands pollute your primary host AI, real projects, or home directory. (applies when: When there are signals of command execution, plugin config, or local writes.) - **Back up your host AI configuration first**: Skill, plugin, and rule files may change the default behavior of Claude/Cursor/Codex. (applies when: When there is a plugin manifest, a Skill, or a host rule entrypoint.) - **After install, verify just one minimal task**: Verify loading, compatibility, output quality, and rollback first, then decide whether to use it deeply. (applies when: When moving from a trial into a real workflow.) ### Exit Plan - **Preserve the pre-install state**: Record the original host config and project state so you can later judge whether it is recoverable. - **Be ready to remove the host plugin / Skill / rule entrypoint**: If behavior is off after the trial install, you can restore the host AI to its pre-trial state. - **Record the install commands and written paths**: Without clear uninstall instructions, you at least need to know which directories or configs to clean up manually. - **If there is no rollback path, do not enter your primary environment**: No rollback is a blocker before continuing; do not proceed on trust or luck. ## What Can Only Be Previewed - Explain who the project fits and what it can do - Demonstrate a typical conversation flow based on project docs - Help the user decide whether it is worth installing or researching further ## What Must Be Verified After Install - Actually installing the Skill, plugin, or CLI - Running scripts, modifying local files, or accessing external services - Verifying real output quality, performance, and compatibility ## Boundary & Risk Decision Card - **Mistaking the pre-install preview for a real run**: The user may overestimate how much configuration, permission, and compatibility verification the project has already done. Mitigation: Clearly separate prompt_preview_can_do from runtime_required. Claim: `clm_0007` inferred 0.45 - **Host AI plugin or Skill rule conflicts**: New rules may change how the user's existing host AI behaves. Mitigation: Inspect the plugin manifest and Skill files before installing, and test in isolation if needed. Evidence: `.claude-plugin/marketplace.json`, `plugin/.claude-plugin/plugin.json` Claim: `clm_0008` supported 0.86 - **Command execution will modify the local environment**: Install commands may write to the user's home directory, the host plugin directory, or project configuration. Mitigation: Run in an isolated environment or a test account first. Evidence: `README.md` Claim: `clm_0009` supported 0.86 - **To confirm**: After a real install, is it compatible with the user's current host AI version?. Why: Compatibility can only be verified in the actual host environment. - **To confirm**: Does the project's output quality meet the user's specific task?. Why: The pre-install preview can only show flow and boundaries; it cannot replace real evaluation. - **To confirm**: Do the install commands require network access, permissions, or global writes?. Why: This affects install risk in both enterprise and personal environments. ## Pre-Work Working Context ### Loading Order - First read how_to_use.host_ai_instruction to establish the boundaries of this pre-install judgment asset. - Read claim_graph_summary to confirm facts come from the Claim/Evidence Graph, not the Human Wiki narrative. - Then read intended_users, capabilities, and quick_start_candidates to judge whether the user is a match. - When you need to carry out a concrete task, check role_skill_index first, then evidence_index. - For real install, file modification, network access, performance, or compatibility questions, turn to risk_card and boundaries.runtime_required. ### Task Routes - **AI Skill / Agent Instruction Asset Library**: Use role_skill_index / evidence_index to help the user pick a usable role, Skill, or workflow first. Boundary: Can be experienced via a pre-install Prompt. Evidence: `.agents/skills/fable-mode/SKILL.md`, `.agents/skills/source-command-hunch-fragile/SKILL.md`, `.claude/skills/fable-mode/SKILL.md`, `plugin/skills/capture/SKILL.md` et al. Claim: `clm_0001` supported 0.86 - **Multi-Host Install and Distribution**: State that this is an after-install capability first, then give a pre-install checklist. Boundary: Must be verified after a real install or run. Evidence: `.claude-plugin/marketplace.json`, `plugin/.claude-plugin/plugin.json` Claim: `clm_0002` supported 0.86 - **Command-Line Startup or Install Flow**: State that this is an after-install capability first, then give a pre-install checklist. Boundary: Must be verified after a real install or run. Evidence: `README.md` Claim: `clm_0003` supported 0.86 ### Context Scale - Total files: 483 - Important-file coverage: 40/483 - Evidence index entries: 80 - Role / Skill entries: 8 ### Handling Insufficient Evidence - **missing_evidence**: State that evidence is insufficient and ask the user for the target file, a README section, or after-install verification records; do not fill in facts. - **out_of_scope_request**: State that the task is beyond the current AI Context Pack's evidence scope and suggest the user check the Human Manual or verify after a real install. - **runtime_request**: Provide a pre-install checklist and command sources, but do not run commands for the user or claim they have been run. - **source_conflict**: Show the conflicting sources side by side, mark them as unverified, and do not force a single version. ## Prompt Recipes ### Fit assessment - Goal: Judge whether this project fits the user's current task. - Expected output: A fit conclusion, key reasons, evidence citations, what can be previewed before install, what must be verified after install, and a next-step recommendation. ```text Based on the AI Context Pack for hunch, ask me 3 necessary questions first, then judge whether it fits my task. The answer must cover: who it fits, what it can do, what it cannot do, whether it is worth installing, and where the evidence comes from. Every project fact must cite evidence_refs, source_paths, or a claim_id. ``` ### Pre-install experience - Goal: Let the user feel the core workflow before installing, while avoiding packaging the preview as real capability or a marketing promise. - Expected output: An experience script with boundary labels, an after-install verification checklist, and a cautious recommendation; with no real-run promises or strong marketing language. ```text Treat hunch as a pre-install experience asset, not an already-installed tool or a real runtime environment. Output exactly four parts: 1. Ask me 3 necessary questions first. 2. Give an "experience script": use the three labels [Previewable before install], [Must verify after install], and [Insufficient evidence] to show how it might guide the workflow. 3. Give an after-install verification checklist: list which capabilities can only be confirmed after a real install, real host loading, and a real project run. 4. Give a cautious recommendation: only "worth researching/trialing further", "add information before deciding", or "not recommended to continue"; do not endorse the project. Hard boundaries: - Do not claim you have installed, run, executed tests, modified files, or produced real results. - Do not write promise-like phrasing such as "auto-adapts", "guarantees passing", "perfect fit", or "strongly recommend installing". - If you describe how it works after install, you must use a conditional such as "if installed successfully and the host loads the Skill correctly, it might...". - The experience script may only be written as "example lines / hypothetical flow": use "might ask / might suggest / might show", not "has written, has generated, has passed, is running, is generating". - Prompt Preview does not hand out install commands; if the user is ready to trial, only prompt them to read Quick Start and the Risk Card first and to verify in an isolated environment. - Every project fact must come from a supported claim, evidence_refs, or source_paths; inferred/unverified items can only be risks or open questions. ``` ### Role / Skill selection - Goal: Pick the best-matching asset from the project's roles or Skills. - Expected output: A list of candidate roles or Skills, each with an applicable scenario, evidence paths, risk boundary, and whether after-install verification is needed. ```text Read role_skill_index and recommend 3-5 of the most relevant roles or Skills for my target task. For each recommendation, state the applicable scenario, likely output, risk boundary, and evidence_refs. ``` ### Risk pre-check - Goal: Identify environment, permission, rule-conflict, and quality risks before installing or adopting. - Expected output: A checklist of environment, permission, dependency, license, host-conflict, quality risk, and unknown items. ```text Based on risk_card, boundaries, and quick_start_candidates, give me a pre-install risk pre-check list. Do not run commands for me; only explain what I should check, why, and what impact a failure would have. ``` ### Host AI kickoff instruction - Goal: Turn the project context into a host AI instruction for the start of a conversation. - Expected output: A pre-work instruction with clear boundaries and clear evidence citations, suitable to copy to a host AI. ```text Based on the AI Context Pack for hunch, generate a pre-work instruction I can paste to my host AI. This instruction must obey not_runtime=true and must not claim the project has been installed, run, or produced real results. ``` ## Role / Skill Index - Indexed 8 role / Skill / project-doc entries. - **fable-mode** (skill): MANDATORY before diagnosing any bug, debugging any failure, reviewing code, or making a non-trivial change — invoke FIRST, before reading code or proposing a cause; do not skip because the task "looks simple". Five-gate rigor protocol scope, evidence, attack, verify, report with per-task playbooks and tripwires. Also fires on "fable mode", "be rigorous", "do this properly", "think hard", and any task where a wrong a… Activation hint: When the user's task is highly relevant to the workflow described by “fable-mode”, use it for a pre-install experience first, then decide whether to install. Evidence: `.agents/skills/fable-mode/SKILL.md` - **source-command-hunch-fragile** (skill): Report the most fragile parts of this codebase, with evidence Activation hint: When the user's task is highly relevant to the workflow described by “source-command-hunch-fragile”, use it for a pre-install experience first, then decide whether to install. Evidence: `.agents/skills/source-command-hunch-fragile/SKILL.md` - **fable-mode** (skill): MANDATORY before diagnosing any bug, debugging any failure, reviewing code, or making a non-trivial change — invoke FIRST, before reading code or proposing a cause; do not skip because the task "looks simple". Five-gate rigor protocol scope, evidence, attack, verify, report with per-task playbooks and tripwires. Also fires on "fable mode", "be rigorous", "do this properly", "think hard", and any task where a wrong a… Activation hint: When the user's task is highly relevant to the workflow described by “fable-mode”, use it for a pre-install experience first, then decide whether to install. Evidence: `.claude/skills/fable-mode/SKILL.md` - **Skill** (skill): Capture an engineering decision into Hunch's graph via a grilling interview topic, rationale, rejected alternatives Activation hint: When the user's task is highly relevant to the workflow described by “Skill”, use it for a pre-install experience first, then decide whether to install. Evidence: `plugin/skills/capture/SKILL.md` - **Skill** (skill): Fix a bug grounded in Hunch past root causes, constraints, blast radius Activation hint: When the user's task is highly relevant to the workflow described by “Skill”, use it for a pre-install experience first, then decide whether to install. Evidence: `plugin/skills/fix/SKILL.md` - **Skill** (skill): Report the most fragile parts of this codebase, with evidence from Hunch Activation hint: When the user's task is highly relevant to the workflow described by “Skill”, use it for a pre-install experience first, then decide whether to install. Evidence: `plugin/skills/fragile/SKILL.md` - **Skill** (skill): Reconcile docs/code with Hunch's decision graph doc≠graph drift , never rewriting prose silently Activation hint: When the user's task is highly relevant to the workflow described by “Skill”, use it for a pre-install experience first, then decide whether to install. Evidence: `plugin/skills/heal/SKILL.md` - **Skill** (skill): Explain why a file or symbol is the way it is, from Hunch's decision graph Activation hint: When the user's task is highly relevant to the workflow described by “Skill”, use it for a pre-install experience first, then decide whether to install. Evidence: `plugin/skills/why/SKILL.md` ## Evidence Index - Indexed 80 evidence entries. - **Hunch** (documentation): Your AI can write code. Hunch makes it remember the consequences. Evidence: `README.md` - **bench — does the skill / the graph actually help?** (documentation): bench — does the skill / the graph actually help? Evidence: `bench/README.md` - **Hunch for VS Code** (documentation): Your repo's engineering memory Hunch https://github.com/davesheffer/hunch , in the editor. One read surface, one write surface, nothing else: Evidence: `vscode-extension/README.md` - **Package** (package_manifest): { "name": "@davesheffer/hunch", "version": "1.7.1", "license": "Apache-2.0", "author": "Dave Sheffer ", "description": "Architectural Conformance for AI-generated code: a git-native graph that deterministically blocks AI changes which break your architecture — the semantic invariants layering, must-reach, dependency direction pattern-SAST can't express — grounded in the decisions and bugs behind each rule, across any MCP assistant Claude Code, Cursor, Copilot, Windsurf, Antigravity, Codex .", "homepage": "https://hunch-pi.vercel.app", "repository": { "type": "git", "url": "git+https://github.com/davesheffer/hunch.git" }, "bugs": { "url": "https://github.com/davesheffer/hunch/issues" }, "typ… Evidence: `package.json` - **Package** (package_manifest): { "name": "hunch-vscode", "displayName": "Hunch", "description": "Your repo's engineering memory, in the editor: see WHY code is the way it is decisions, invariants, bug history , and capture what you decide — one command each.", "version": "0.14.0", "publisher": "davesheffer", "license": "MIT", "author": "Dave Sheffer ", "icon": "media/icon.png", "homepage": "https://hunch-pi.vercel.app", "repository": { "type": "git", "url": "git+https://github.com/davesheffer/hunch.git", "directory": "vscode-extension" }, "bugs": { "url": "https://github.com/davesheffer/hunch/issues" }, "galleryBanner": { "color": " 16150f", "theme": "dark" }, "keywords": "claude-code", "mcp", "engineering-memory", "know… Evidence: `vscode-extension/package.json` - **AGENTS.md** (documentation): This repo has Hunch — a curated graph of why the code is the way it is decisions, bug history, invariants . It currently holds 101 decisions, 1 bugs, 8 constraints, 16 components, 3 policies . Evidence: `AGENTS.md` - **CLAUDE.md** (documentation): This file provides guidance to Claude Code claude.ai/code when working with code in this repository. Evidence: `CLAUDE.md` - **Contributing to Hunch** (documentation): Thanks for helping build the Engineering Memory OS. Evidence: `CONTRIBUTING.md` - **Fable Mode** (skill_instruction): Each gate exists to kill one specific failure mode. The gates shape the work; they are invisible in the output — only Gate 5 is what the user sees. Never narrate "now entering Gate 2." A failed gate sends you backward, never forward. Evidence: `.agents/skills/fable-mode/SKILL.md` - **source-command-hunch-fragile** (skill_instruction): Use this skill when the user asks to run the migrated source command hunch-fragile . Evidence: `.agents/skills/source-command-hunch-fragile/SKILL.md` - **Fable Mode** (skill_instruction): Each gate exists to kill one specific failure mode. The gates shape the work; they are invisible in the output — only Gate 5 is what the user sees. Never narrate "now entering Gate 2." A failed gate sends you backward, never forward. Evidence: `.claude/skills/fable-mode/SKILL.md` - **Skill** (skill_instruction): Capture the decision for $ARGUMENTS into Hunch's graph. Evidence: `plugin/skills/capture/SKILL.md` - **Skill** (skill_instruction): Follow the Hunch-grounded workflow — do NOT skip the memory lookups: 1. hunch bug lineage "$ARGUMENTS" — has this class of bug happened before? what was the root cause and the fix? 2. Identify the suspect symbol/file, then hunch get dependents to learn the blast radius. 3. hunch check constraints — list invariants you must preserve. 4. Propose a fix that honors past root causes AND constraints. Apply it and run the tests. 5. If the fix encodes a non-trivial choice, hunch record decision ... so the next session is grounded in it. Evidence: `plugin/skills/fix/SKILL.md` - **Skill** (skill_instruction): Ask Hunch for the fragility ranking run hunch fragile or query Hunch , then produce a fragility report with evidence : the specific files/functions, the bug history behind them, their churn and fan-in, and any missing guards. Avoid generic advice — every claim must cite a Hunch record or metric. Evidence: `plugin/skills/fragile/SKILL.md` - **Skill** (skill_instruction): Reconcile decision-grounding drift for $ARGUMENTS or the whole repo . Evidence: `plugin/skills/heal/SKILL.md` - **Skill** (skill_instruction): Use the hunch why MCP tool on $ARGUMENTS a file path or symbol name . Evidence: `plugin/skills/why/SKILL.md` - **Marketplace** (structured_config): { "name": "hunch", "owner": { "name": "Dave Sheffer", "email": "dave.sheffer1@gmail.com" }, "description": "Hunch — Engineering Memory OS: a git-native reasoning graph of decisions, bugs, and invariants, enforced at reasoning time.", "plugins": { "name": "hunch", "source": "./plugin", "description": "The hunch MCP tools why, constraints, blast radius, merge verdict, capture plus /hunch:capture, /hunch:heal, /hunch:why, /hunch:fix, /hunch:fragile. Requires Node 22.13+; runs npx @davesheffer/hunch mcp per project.", "author": { "name": "Dave Sheffer", "email": "dave.sheffer1@gmail.com" }, "homepage": "https://hunch-pi.vercel.app", "repository": "https://github.com/davesheffer/hunch", "license… Evidence: `.claude-plugin/marketplace.json` - **Plugin** (structured_config): { "name": "hunch", "displayName": "Hunch — Engineering Memory OS", "description": "A git-native reasoning graph of decisions, bugs, and invariants for your codebase — queried and enforced at reasoning time. Adds the hunch MCP tools plus /hunch:capture, /hunch:heal, /hunch:why, /hunch:fix, /hunch:fragile.", "author": { "name": "Dave Sheffer", "email": "dave.sheffer1@gmail.com", "url": "https://github.com/davesheffer" }, "homepage": "https://hunch-pi.vercel.app", "repository": "https://github.com/davesheffer/hunch", "license": "Apache-2.0", "keywords": "engineering-memory", "architectural-conformance", "decisions", "invariants", "mcp", "knowledge-graph" , "mcpServers": "./.mcp.json", "skills"… Evidence: `plugin/.claude-plugin/plugin.json` - **License** (source_file): Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ Evidence: `LICENSE` - **License** (source_file): Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files the "Software" , to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: Evidence: `vscode-extension/LICENSE` - **Hunch Cookbook** (documentation): Practical recipes for running Hunch day to day. Each recipe is copy-paste-able and states what you should observe when it works. Evidence: `docs/cookbook.md` - **Index** (source_file): import { existsSync, readFileSync, writeFileSync, mkdirSync, realpathSync } from "node:fs"; import { execFileSync, spawnSync } from "node:child process"; import { join, relative, dirname, basename, resolve, isAbsolute } from "node:path"; import { fileURLToPath } from "node:url"; import { Command } from "commander"; import { hunchPaths, hunchPathsForDir, findRoot, toPosixTarget } from "../core/paths.js"; import { writeFileAtomic } from "../core/io.js"; import { looksLikeCorrection, CORRECTION NUDGE } from "../core/correction.js"; import { HUNCH VERSION } from "../core/version.js"; import { HunchStore } from "../store/hunchStore.js"; import { JsonStore } from "../store/jsonStore.js"; import {… Evidence: `src/cli/index.ts` - **Behaviorproof** (source_file): import { basename } from "node:path"; import { shortHash } from "../core/ids.js"; import { headSha } from "../extractors/git.js"; import { evaluateExecutableBehaviorPolicy } from "./behaviorEvaluator.js"; import { canonicalHash, policySemanticHash, proofEvaluationHash, proofId, proofPlanContentHash } from "./canonical.js"; import type { PolicyRepository } from "./repository.js"; import { BEHAVIOR MUTATION ENGINE, BEHAVIOR POLICY EVALUATOR, MutationReceiptSchema, PolicyProofSchema, ProofPlanSchema, ReplayReceiptSchema, type EvaluationSummary, type MutationReceipt, type PolicyEvaluation, type PolicyProof, type PolicySpec, type ProofPlan, type ReplayReceipt, } from "./schema.js"; ⋮---- functio… Evidence: `src/constitution/behaviorProof.ts` - **Bootstrap** (source_file): import { basename } from "node:path"; import { shortHash } from "../core/ids.js"; import type { Decision } from "../core/types.js"; import type { HunchStore } from "../store/hunchStore.js"; import { canonicalHash } from "./canonical.js"; import { compileDecisionRecord } from "./compiler.js"; import type { PolicyRepository } from "./repository.js"; import { EvidenceEventSchema, type EvidenceEvent, type PolicySpec } from "./schema.js"; ⋮---- export interface BootstrapCandidate { evidence: EvidenceEvent; policy: PolicySpec; } ⋮---- export interface BootstrapReport { scanned: number; eligible: number; compiled: BootstrapCandidate ; covered: number; deferred: number; uncompilable: number; confli… Evidence: `src/constitution/bootstrap.ts` - **Compiler** (source_file): import type { Decision, ConformancePredicate } from "../core/types.js"; import type { HunchStore } from "../store/hunchStore.js"; import { policyId } from "./canonical.js"; import { POLICY IR VERSION, PolicySpecSchema, type CandidateContext, type DataClass, type PolicyAssertion, type PolicySelector, type PolicySpec } from "./schema.js"; ⋮---- function selector ref: string : PolicySelector ⋮---- function compileAssertion predicate: ConformancePredicate, through?: string : PolicyAssertion ⋮---- export interface CompilePolicyOptions { through?: string; private?: boolean; now?: string; } ⋮---- export function compileDecisionRecord store: HunchStore, source: Decision, isPrivate: boolean, opts: O… Evidence: `src/constitution/compiler.ts` - **G3Conformance** (source_file): import { spawnSync } from "node:child process"; import { mkdtempSync, readFileSync, rmSync, writeFileSync } from "node:fs"; import { tmpdir } from "node:os"; import { join } from "node:path"; import { pathToFileURL } from "node:url"; import { canonicalHash } from "./canonical.js"; import type { G3Plan, AdapterConformanceReceipt } from "./g3.js"; import { compileAdapterConformanceReceipt } from "./g3.js"; import { NODE TEST REPORTER SOURCE, exactNodeTestPattern, nodeTestIsolationFlag, nodeTestReporterEvents } from "./nodeTestEvidence.js"; ⋮---- export function g3ConformanceSourceHash root: string : string ⋮---- function sameClients left: readonly string , right: readonly string : boolean ⋮--… Evidence: `src/constitution/g3Conformance.ts` - **Lifecycle** (source_file): import { assertCompositionBinding, exceptionScopeIsNarrower, oppositeExceptionAssertions, policyProofHash } from "./composition.js"; import { assessHistoryDispositions } from "./disposition.js"; import { type HistoryDisposition, type PolicyProof, type PolicySpec, type ProofClass } from "./schema.js"; import { evaluatorForPolicy, mutationEngineForPolicy } from "./policyRuntime.js"; import { executableBehaviorAttestationError } from "./behaviorAttestationBinding.js"; import type { G2BehaviorAttestation } from "./g2BehaviorAttestation.js"; ⋮---- function isHumanActor actor: string : boolean ⋮---- export function blockingEvidenceError proof: PolicyProof, dispositions: HistoryDisposition = : str… Evidence: `src/constitution/lifecycle.ts` - **Plan** (source_file): import { basename } from "node:path"; import { shortHash } from "../core/ids.js"; import { firstCommitForFile, headSha, revExists, revParse } from "../extractors/git.js"; import type { HunchStore } from "../store/hunchStore.js"; import { canonicalHash, policySemanticHash } from "./canonical.js"; import { policyCompositionBinding, policyProofHash } from "./composition.js"; import { graphSnapshot, mutationOperatorForPolicy, selectedPolicyForComposition } from "./evaluator.js"; import type { PolicyRepository } from "./repository.js"; import { createExecutableBehaviorProofPlan } from "./behaviorProof.js"; import { POLICY EVALUATOR, MUTATION ENGINE, ProofPlanSchema, type EvidenceEvent, type Poli… Evidence: `src/constitution/plan.ts` - **Proof** (source_file): import { canonicalHash, proofEvaluationHash, proofId } from "./canonical.js"; import { assertCompositionBinding, policyCompositionBinding, policyProofHash } from "./composition.js"; import { evaluateCompositePolicyOnSnapshot, evaluatePolicyOnSnapshot, graphSnapshot, mutationOperatorForPolicy, selectedPolicyForComposition, type GraphSnapshot } from "./evaluator.js"; import { runMutationHarness } from "./mutation.js"; import { replayProofPlan } from "./replay.js"; import { POLICY EVALUATOR, MUTATION ENGINE, PolicyProofSchema, type EvaluationSummary, type PolicyEvaluation, type PolicyProof, type PolicySpec, type ProofPlan, type ProofClass, type ReplayReceipt, } from "./schema.js"; import type… Evidence: `src/constitution/proof.ts` - **Schema** (source_file): import { z } from "zod"; import { ProvenanceSchema } from "../core/types.js"; ⋮---- export type DataClass = z.infer ; ⋮---- export type StructuralSymbolRef = z.infer ; ⋮---- export type StructuralCallRef = z.infer ; ⋮---- export type StructuralImportRef = z.infer ; ⋮---- export type StructuralDelta = z.infer ; ⋮---- export type CandidateAlternative = z.infer ; ⋮---- export type PolicyScope = z.infer ; ⋮---- export type CandidateContext = z.infer ; ⋮---- export type EvidenceEvent = z.infer ; ⋮---- export type EvidenceImportItem = z.infer ; ⋮---- export type EvidenceImport = z.infer ; ⋮---- export type PolicyState = z.infer ; ⋮---- export type PolicySelector = z.infer ; ⋮---- export type Poli… Evidence: `src/constitution/schema.ts` - **Config** (source_file): import { existsSync, readFileSync, writeFileSync, mkdirSync } from "node:fs"; import { dirname } from "node:path"; import type { HunchPaths } from "./paths.js"; ⋮---- export type Firmness = "off" "advisory" "firm" "strict"; ⋮---- export interface HunchConfig { firmness: Firmness; } ⋮---- function defaults : HunchConfig ⋮---- export function isFirmness v: unknown : v is Firmness ⋮---- export function readConfig paths: HunchPaths : HunchConfig ⋮---- export function writeConfig paths: HunchPaths, patch: Partial : HunchConfig Evidence: `src/core/config.ts` - **Conformance** (source_file): import type { HunchStore } from "../store/hunchStore.js"; import type { Decision, ConformancePredicate } from "./types.js"; ⋮---- export interface ConformanceResult { decision: string; title: string; assert: string; subject: string; object?: string; satisfied: boolean; detail: string; } ⋮---- function resolveSymbol store: HunchStore, ref: string : ⋮---- function reaches store: HunchStore, id: string, transitive: boolean : Set ⋮---- function evalPredicate store: HunchStore, d: Decision, p: ConformancePredicate : ConformanceResult ⋮---- / Check every in-force decision's conformance predicates against the CURRENT graph. .satisfied === false means the code drifted from the recorded intent. Dete… Evidence: `src/core/conformance.ts` - **Pipeline** (source_file): import { mkdirSync, readFileSync, readdirSync, rmSync, statSync, writeFileSync } from "node:fs"; import { tmpdir } from "node:os"; import { join } from "node:path"; ⋮---- export type Firmness = "off" "advisory" "firm" "strict"; ⋮---- export interface PipelineState { turn: number; soulInjected: boolean; blocks: number; domains: Record ; editedFiles: string ; verifyAfterEdit: boolean; } ⋮---- export const emptyState = : PipelineState = ⋮---- export interface DomainProfile { paths: RegExp; verify: RegExp; } ⋮---- export function isProductPath p: string : boolean ⋮---- export function classifyDomains path: string, profiles = DEFAULT PROFILES : string ⋮---- function verifyPattern state: Pipeline… Evidence: `src/core/pipeline.ts` - **Strictgate** (source_file): export function isHumanConfirmed source: string undefined : boolean ⋮---- export interface StrictConstraint { severity?: string; provenance?: { confidence?: number; source?: string }; } ⋮---- export function isStrictBlocker c: StrictConstraint, stale: boolean : boolean ⋮---- export type VetoTier = "dep" "symbol" "pattern" "semantic"; ⋮---- export function isVetoBlocker d: { status?: string; superseded by?: string null }, tw: { provenance?: { source?: string } }, tier: VetoTier, stale: boolean, : boolean Evidence: `src/core/strictgate.ts` - **Types** (source_file): import { z } from "zod"; ⋮---- export type Provenance = z.infer ; ⋮---- export type Component = z.infer ; ⋮---- export type Edge = z.infer ; ⋮---- export type SymbolMetrics = z.infer ; ⋮---- export type Symbol = z.infer ; ⋮---- export type RetiredSignal = z.infer ; ⋮---- export type RejectedTripwire = z.infer ; ⋮---- export type ConformancePredicate = z.infer ; ⋮---- export type Decision = z.infer ; ⋮---- export type BugLineage = z.infer ; ⋮---- export type Bug = z.infer ; ⋮---- // Bi-temporal VALID-TIME: a constraint can be RETIRED without deletion, so // "what invariants were in force as of commit X?" stays answerable. valid to // null = still active. A retired constraint is excluded from… Evidence: `src/core/types.ts` - **Version** (source_file): import { readFileSync } from "node:fs"; import { fileURLToPath } from "node:url"; import { dirname, join } from "node:path"; Evidence: `src/core/version.ts` - **Claudeconfig** (source_file): import { homedir } from "node:os"; import { join } from "node:path"; import { existsSync, readFileSync, copyFileSync } from "node:fs"; import { writeFileAtomic } from "../core/io.js"; ⋮---- export function claudeConfigPath : string ⋮---- export interface HealedGroup { realPath: string; casings: string ; servers: string ; } ⋮---- export interface HealResult { platform: NodeJS.Platform; applicable: boolean; file: string; backup?: string; changed: boolean; groups: HealedGroup ; } ⋮---- export interface HealOptions { file?: string; platform?: NodeJS.Platform; } ⋮---- type ProjectBlock = Record & { mcpServers?: Record ; enabledMcpjsonServers?: unknown; disabledMcpjsonServers?: unknown; }; ⋮----… Evidence: `src/integrations/claudeConfig.ts` - **Claudemd** (source_file): import { readFileSync, writeFileSync, existsSync, mkdirSync } from "node:fs"; import { join, dirname } from "node:path"; import type { HunchStore } from "../store/hunchStore.js"; import { wikiSummary } from "../wiki/wiki.js"; import { PolicyRepository } from "../constitution/repository.js"; ⋮---- export function renderHunchSection store: HunchStore, root?: string : string ⋮---- export function upsertSection file: string, section: string, fallbackTitle: string : string ⋮---- // partial/corrupt markers: strip stray marker lines, then append ONE clean section. ⋮---- export function updateClaudeMd root: string, store: HunchStore : string ⋮---- function sev s: string : number Evidence: `src/integrations/claudemd.ts` - **Hooks** (source_file): import { readFileSync, writeFileSync, existsSync, chmodSync, mkdirSync } from "node:fs"; import { join, isAbsolute } from "node:path"; import { hooksDir } from "../extractors/git.js"; ⋮---- function block invocation: string, opts: ⋮---- export interface HookInstall { path: string; action: "created" "appended" "updated" "unchanged"; } ⋮---- export function installPostCommitHook root: string, invocation: string, opts: ⋮---- function escapeRe s: string : string ⋮---- export function installPreCommitHook root: string, invocation: string, strict = false : HookInstall Evidence: `src/integrations/hooks.ts` - **Mergedriver** (source_file): import { execFileSync } from "node:child process"; import { readFileSync, existsSync } from "node:fs"; import { join } from "node:path"; import { writeFileAtomic } from "../core/io.js"; ⋮---- export function installMergeDriver root: string, invShell: string : Evidence: `src/integrations/mergeDriver.ts` - **Providers** (source_file): import { readFileSync, writeFileSync, existsSync, mkdirSync } from "node:fs"; import { homedir } from "node:os"; import { join, dirname } from "node:path"; import type { HunchStore } from "../store/hunchStore.js"; import type { Invocation } from "./scaffold.js"; import type { HookProvider } from "../core/agenthook.js"; import { renderHunchSection, upsertSection, updateClaudeMd } from "./claudemd.js"; import { isGitCleanPath } from "../extractors/git.js"; ⋮---- function stripJsonc s: string : string ⋮---- function dropTrailingCommas s: string : string ⋮---- if s j === "}" s j === " " continue; // trailing comma → drop ⋮---- / Read a JSON/JSONC object. Returns {} only for an ABSENT or empty f… Evidence: `src/integrations/providers.ts` - **Scaffold** (source_file): import { readFileSync, writeFileSync, existsSync, mkdirSync } from "node:fs"; import { join, dirname } from "node:path"; ⋮---- export interface Invocation { command: string; args: string ; } ⋮---- export function writeMcpJson root: string, inv: Invocation : string ⋮---- export interface ClaudeHookInstall { path: string; action: "created" "updated" "unchanged"; } ⋮---- interface HookEntry { matcher?: string; hooks?: Array ; } ⋮---- function isHunchHook entry: HookEntry : boolean ⋮---- / Install the Claude Code AGENT hooks into .claude/settings.json so the agent is grounded in Hunch automatically not by remembering to call the tools : - PreToolUse Edit Write MultiEdit → inject the relevant Hu… Evidence: `src/integrations/scaffold.ts` - **Server** (source_file): import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"; import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js"; import { z } from "zod"; import { hunchPaths, findRoot, toPosixTarget } from "../core/paths.js"; import { HunchStore } from "../store/hunchStore.js"; import { selectEmbedder } from "../store/embedder.js"; import { decisionId } from "../core/ids.js"; import { buildCorrectionConstraint } from "../core/correction.js"; import { knownRepoDeps } from "../synthesis/tripwires.js"; import { refreshExistingGrounding } from "../integrations/providers.js"; import { revParse, asOfDate, revExists, lastChangeDate, rangeFiles, rangeDiff, commitFiles, commitD… Evidence: `src/mcp/server.ts` - **Compact** (source_file): import type { Bug, Constraint, Decision, EntityKind } from "../core/types.js"; ⋮---- export interface CompactionCandidate { kind: EntityKind; id: string; title: string; reason: string; } ⋮---- export interface CompactionPlan { remove: CompactionCandidate ; considered: number; } ⋮---- export interface CompactionInput { decisions: Decision ; bugs: Bug ; constraints: Constraint ; } ⋮---- export interface CompactionOpts { now: number; maxAgeDays?: number; minConfidence?: number; } ⋮---- export function planCompaction input: CompactionInput, opts: CompactionOpts : CompactionPlan ⋮---- const ageDays = iso: string : number = ⋮---- const decReason = d: Decision : string null = const bugReason = b:… Evidence: `src/store/compact.ts` - **Db** (source_file): import { createRequire } from "node:module"; import { mkdirSync } from "node:fs"; import { dirname } from "node:path"; import { SCHEMA SQL } from "./schema.js"; ⋮---- function loadSqlite : typeof import "node:sqlite" ⋮---- export type DB = import "node:sqlite" .DatabaseSync; ⋮---- export function openDb sqlitePath: string : DB ⋮---- export function openMemoryDb : DB ⋮---- export function withTx db: DB, fn: = T : T Evidence: `src/store/db.ts` - **Embedder** (source_file): import { createRequire } from "node:module"; ⋮---- export interface Embedder { embed texts: string : Promise ; readonly dim: number; readonly id: string; } ⋮---- embed texts: string : Promise ; ⋮---- function installedPackage : string null ⋮---- export class TransformersEmbedder implements Embedder ⋮---- private load ⋮---- async embed texts: string : Promise ⋮---- export class StubEmbedder implements Embedder ⋮---- constructor readonly dim = 32 ⋮---- export async function selectEmbedder : Promise Evidence: `src/store/embedder.ts` - **Hunchstore** (source_file): import { resolve, join } from "node:path"; import { existsSync, readFileSync } from "node:fs"; import { toPosixTarget, hunchPathsForDir, type HunchPaths } from "../core/paths.js"; import { ENTITY KINDS, type Component, type Constraint, type Bug, type Decision, type Symbol, type Edge, type RejectedTripwire, type EntityKind, type EntityFor } from "../core/types.js"; import { openDb, withTx, type DB } from "./db.js"; import { RESET SQL, embedHash } from "./schema.js"; import { selectEmbedder, type Embedder } from "./embedder.js"; import { JsonStore } from "./jsonStore.js"; import { gitCommonDir } from "../extractors/git.js"; import { pathMatchesGlob } from "../core/glob.js"; import { currentFo… Evidence: `src/store/hunchStore.ts` - **Merge** (source_file): type Rec = Record ; ⋮---- export interface MergeResult { text: string; conflict: boolean; } ⋮---- export function mergeHunchJson baseText: string, oursText: string, theirsText: string : MergeResult ⋮---- if merged.length === 0 return { text: "", conflict: false }; // both deleted the record // A single-record file that produced 1 record means a side rewrote the id // a logical rename . Don't silently drop one — fall back so git surfaces it. ⋮---- / Three-way merge of record arrays keyed by id . Additions on either side are kept; a record changed on one side only takes that side; a delete is honored only if the other side left the record unchanged a modification beats a delete ; a both-sides… Evidence: `src/store/merge.ts` - **Schema** (source_file): import { shortHash } from "../core/ids.js"; ⋮---- export function embedHash title: string, body: string : string Evidence: `src/store/schema.ts` - **Providers** (source_file): import { symbolSignals, bugsForSymbol, type Hunch } from "./hunchData.js"; ⋮---- type GetHunch = = Hunch null; type RelPath = file: string = string; ⋮---- export class HunchHoverProvider implements vscode.HoverProvider ⋮---- constructor private getHunch: GetHunch, private rel: RelPath ⋮---- provideHover doc: vscode.TextDocument, pos: vscode.Position : vscode.Hover undefined Evidence: `vscode-extension/src/providers.ts` - **Security Policy** (documentation): Email dave.sheffer1@gmail.com or open a private security advisory https://github.com/davesheffer/hunch/security/advisories/new . Please do not file public issues for undisclosed vulnerabilities. Expect an acknowledgement within a few days. Evidence: `SECURITY.md` - **Architectural Conformance — prevent + catch benchmark** (documentation): Architectural Conformance — prevent + catch benchmark Evidence: `bench/architectural-conformance.md` - **We let the AI audit its own memory tool. It was using 4 of 19 tools.** (documentation): We let the AI audit its own memory tool. It was using 4 of 19 tools. Evidence: `blog/agent-audits-its-own-memory.md` - **AI ignores your architecture rules — even at the frontier. We measured it.** (documentation): AI ignores your architecture rules — even at the frontier. We measured it. Evidence: `blog/ai-ignores-your-architecture.md` - **Fable Mode — Backend playbook** (documentation): APIs, services, data, contracts, migrations. The gates, sharpened for state you can corrupt and callers you can't see. Evidence: `.agents/skills/fable-mode/references/backend.md` - **Fable Mode — Frontend playbook** (documentation): UI, components, styling, browser behavior. The gates, sharpened for pixels. Evidence: `.agents/skills/fable-mode/references/frontend.md` - **Fable Mode — Infra / DevOps playbook** (documentation): Fable Mode — Infra / DevOps playbook Evidence: `.agents/skills/fable-mode/references/infra.md` - **Fable Mode — QA / test playbook** (documentation): Writing tests, fixing tests, building test plans. The prime directive: a test's value is its ability to FAIL. A test you never saw fail is a hope, not a test. Evidence: `.agents/skills/fable-mode/references/qa.md` - **Fable Mode — Verdict playbook** (documentation): Design decisions, red teams, "which option is right." Not for execution work — this playbook manufactures a CALL, and its byproducts are the record: real rejected alternatives, evidenced attacks, flip conditions. Evidence: `.agents/skills/fable-mode/references/verdict.md` - **Capture** (documentation): Capture the decision for $ARGUMENTS into Hunch's graph. Evidence: `.claude/commands/capture.md` - The remaining 20 evidence entries are in `AI_CONTEXT_PACK.json` or `EVIDENCE_INDEX.json`. ## Rules the Host AI Must Follow - **Treat this asset as pre-work context, not a runtime environment.**: The AI Context Pack contains only an evidence-backed understanding of the project, not the project's executable state. Evidence: `README.md`, `bench/README.md`, `vscode-extension/README.md` - **When answering the user, distinguish what can be previewed from what can only be verified after install.**: The consumer value of the pre-install experience comes from reducing bad installs and misjudgments, not from pretending to be a real run. Evidence: `README.md`, `bench/README.md`, `vscode-extension/README.md` ## Questions the User Should Answer First - Which host AI or local environment do you plan to use it in? - Do you just want to experience the workflow first, or are you ready to actually install? - What matters most to you: install cost, output quality, or conflicts with your existing rules? ## Acceptance Checks - Every capability claim can be traced back to a file path in evidence_refs. - AI_CONTEXT_PACK.md does not package previews as a real run. - The user can understand who it fits, what it can do, how to start, and the risk boundaries within 3 minutes. --- ## Doramagic Context Augmentation The following sections strengthen the repository context for a host AI. Human Manual data is a reading route, and pitfall notes become operating constraints. ## Human Manual Outline Usage rule: this is only a reading route and salience signal, not factual authority. Concrete claims must still return to repo evidence or Claim Graph. Host AI hard rules: - Do not treat page titles, section order, summaries, or importance values as factual project evidence. - When explaining the Human Manual outline, state that it is only a reading route or salience signal. - Capability, installation, compatibility, runtime state, and risk claims must cite repo evidence, source paths, or Claim Graph. - **Overview, Repository Layout & System Architecture**: importance `high` - source_paths: README.md, src/cli/index.ts, src/core/pipeline.ts, src/core/types.ts, src/core/config.ts - **Change Gate, Constitution & Policy Enforcement**: importance `high` - source_paths: src/core/strictgate.ts, src/core/conformance.ts, src/constitution/bootstrap.ts, src/constitution/compiler.ts, src/constitution/plan.ts - **Memory Graph, Storage & Data Flow**: importance `high` - source_paths: src/store/hunchStore.ts, src/store/db.ts, src/store/schema.ts, src/store/embedder.ts, src/store/compact.ts - **Agent Integrations, Lifecycle Hooks & Synthesis Providers**: importance `high` - source_paths: src/integrations/providers.ts, src/integrations/scaffold.ts, src/integrations/claudeConfig.ts, src/integrations/claudemd.ts, src/integrations/hooks.ts ## Repo Inspection Evidence - repo_clone_verified: true - repo_inspection_verified: true - repo_commit: `62842da2bf949d50099eb16b16322428cfa6f983` - inspected_files: `README.md`, `package.json`, `docs/cookbook.md`, `docs/evidence/constitution-phase-3-clean-install.json`, `src/cli/index.ts`, `src/cli/invocation.ts`, `src/cli/preflight.ts`, `src/constitution/adapters.ts`, `src/constitution/behaviorAttestationBinding.ts`, `src/constitution/behaviorEvaluator.ts`, `src/constitution/behaviorProof.ts`, `src/constitution/behaviorWorkspace.ts`, `src/constitution/bootstrap.ts`, `src/constitution/canonical.ts`, `src/constitution/card.ts`, `src/constitution/compiler.ts`, `src/constitution/composition.ts`, `src/constitution/corpus.ts`, `src/constitution/delta.ts`, `src/constitution/disposition.ts` Host AI hard rules: - Without repo_clone_verified=true, do not claim that the source code has been read. - Without repo_inspection_verified=true, do not write README, docs, or package-file conclusions as facts. - Without quick_start_verified=true, do not claim that the Quick Start path has run successfully. ## Doramagic Pitfall Constraints These rules come from Doramagic discovery, validation, or compilation findings. The host AI must treat them as operating constraints, not background notes. ### Constraint 1: Capability evidence risk requires verification - Trigger: README/documentation is current enough for a first validation pass. - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: capability.assumptions | https://github.com/davesheffer/hunch - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it. ### Constraint 2: Security or permission risk requires verification - Trigger: no_demo - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: downstream_validation.risk_items | https://github.com/davesheffer/hunch - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it. ### Constraint 3: Security or permission risk requires verification - Trigger: no_demo - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: risks.scoring_risks | https://github.com/davesheffer/hunch - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it. ### Constraint 4: Maintenance risk requires verification - Trigger: issue_or_pr_quality=unknown。 - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: evidence.maintainer_signals | https://github.com/davesheffer/hunch - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it. ### Constraint 5: Maintenance risk requires verification - Trigger: release_recency=unknown。 - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: evidence.maintainer_signals | https://github.com/davesheffer/hunch - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it.