docs: catch wiki up to v9.2 The wiki had drifted across releases. Tool/preset counts, token-savings percentages, LIP version, and review-check counts all referenced older numbers. Several v9.1/v9.2 features were missing entirely. Numbers refreshed against running binary (`ckb mcp --list-presets`): - Preset sizes: core 25, review 42, refactor 42, federation 46, docs 34, ops 43, full 110 (was 14/19/19/28/20/25/80+) - Full preset: ~17k tokens / 77% savings (was ~9k / 83%) - LIP: v2.3+ (was v2.0) - `ckb review`: 21 checks (was 20 — `unwired` was missing) Documented features that had no wiki coverage: - v9.2 MCP tools `analyzeOutgoingImpact`, `symbolExists`, `renderArchitecture` - Cartographer 3.0.0 graph rendering (Mermaid + Graphviz DOT) - LIP v2.3 outgoing-impact and semantic-coupling enrichment - Cartographer.md was authored locally but never committed

SimplyLiz committed Apr 25, 2026

docs(v9.0.0): LIP v2.0 integration + correct setup instructions - Home.md: What's New in v9.0 section with LIP semantic layer bullets - Hybrid-Retrieval.md: correct LIP daemon startup command, add LIP_EMBEDDING_URL/LIP_EMBEDDING_MODEL env vars, note context-limit workaround for large files, update doctor output format - MCP-Tools.md: mark tools enhanced by LIP (✦) with link to lip docs - Quick-Start.md: two-mode LIP setup (graph-only vs. with embeddings) - Architecture.md, Contributing.md, Practical-Limits.md: minor v9 refs Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

SimplyLiz committed Apr 13, 2026

docs: add Unwired-Modules.md wiki page Full documentation for findUnwiredModules: algorithm (import-graph BFS + call-graph BFS), CLI/HTTP/MCP reference, PR review check, real-world example from LLMRouter, entrypoint detection sources, exclusions, per-language support table. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Apr 2, 2026

docs: add findUnwiredModules to wiki docs - Features.md: add unwired detection to Technical Debt section - Code-Review.md: add unwired check to check table, update 20→21 checks - API-Reference.md: add GET /unwired endpoint documentation Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Apr 1, 2026

docs: add A2A protocol documentation and cross-references - New page: A2A-Integration.md — full A2A protocol guide with setup, agent discovery, protocol bindings, skill reference, health/index status, push notifications, auth, and integration examples - Home.md: add A2A/agent platforms to "Works With Your Tools" table - Architecture.md: add A2A Server to interface layer diagram - Features.md: add "Agent-to-Agent Orchestration" scenario section - Integration-Guide.md: add "Multi-Agent Systems (A2A)" ideas table - Quick-Start.md: add A2A server startup section Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Apr 1, 2026

docs: update tool counts across all wiki pages for v8.3 Update stale tool/preset counts in 8 pages: - core: 14→24, review: 27→40, refactor: 33→39, full: 92→96 - "80+ tools" / "90+ tools" → "96 tools" everywhere - Token estimates updated to match new preset sizes Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Mar 27, 2026

docs: update wiki for v8.3 — new MCP tools, --recommend, tool counts Compliance-Audit.md: Add --recommend flag to CLI Reference. MCP-Tools.md: Add auditCompliance, listSymbols, getSymbolGraph tool entries with full parameter/return documentation. Update tool count from 90+ to 96. Features.md: Update preset tool counts (core 24, review 40, refactor 39, full 96). Add listSymbols, getSymbolGraph, auditCompliance to compound operations table. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Mar 27, 2026

docs: update check count from 19 to 20 across wiki pages Review now has 20 checks (split was added as the 20th). Update all references in Integration-Guide, MCP-Tools, Performance, Quality-Gates, User-Guide, and Workflow-Examples. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Mar 25, 2026

docs: fix score calculation in Code-Review — floor is 0 not 20, add per-rule cap - Score floor is 0, not 20 (matches review.go:1166-1168) - Document per-rule deduction cap of 10 (review.go:1107) — prevents a single noisy rule from consuming its check's entire budget - Clarify three-level cap system: per-rule (10), per-check (20), total (80) Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Mar 25, 2026

docs: rewrite Compliance-Audit page to match actual implementation - Fix command: `ckb review --format=compliance` → `ckb audit compliance --framework=...` - Fix flags: `--compliance-frameworks` → `--framework`, `--fail-on=high` → `--fail-on=error` - Update framework list: remove 6 fictional (NIST CSF, FedRAMP, FISMA, CIS Controls, OWASP Top 10, ISO 21434), add 6 actual (EU AI Act, EU CRA, FDA 21 CFR 11, ISO 27701, ISO 26262, SBOM/SLSA) - Add real cross-framework mapping table (16 categories with CWE IDs) - Add actual JSON output format matching ComplianceReport struct - Add relationship section explaining `ckb review` vs `ckb audit compliance` - Fix Features.md commands and framework table to match - Add Compliance Audit section to CI-CD-Integration.md Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Mar 25, 2026

feat: Add Compliance-Audit wiki page, update Home and Features - New Compliance-Audit.md with full documentation: 20 frameworks, 126 checks, cross-framework mapping, check details for GDPR/PCI DSS/ HIPAA/SOC 2/safety, configuration, CI/CD integration, CLI reference - Home.md: added compliance audit entry - Features.md: added compliance audit section with framework overview Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SimplyLiz committed Mar 25, 2026

docs: Update review docs for v8.3 — 19 checks, new analyzers - 17→19 checks across all wiki pages - Add comment-drift and format-consistency to checks table - Update blast-radius: informational mode when no threshold set - Add check list to Workflow-Examples

SimplyLiz committed Mar 20, 2026

docs: Add structured PR triage prompt to Prompt Cookbook System prompt for large PR review with token budget enforcement. Uses ckb review output to build a 4-tier reading list, enforces file read caps, and replaces naive file reading with CKB tool calls.

SimplyLiz committed Mar 19, 2026

docs: Update review docs for v8.3 — 17 checks, new analyzers, architecture diagram - 14→17 checks across all pages (Code-Review, CI-CD, Quality-Gates, Features, Home, User-Guide, Workflow-Examples, MCP-Tools) - Add dead-code, test-gaps, blast-radius to check table and examples - Add --staged, --scope, positional arg invocation modes - Add max-fanout, dead-code-confidence, test-gap-lines config/flags - Add architecture SVG reference in Code-Review page - Update concurrency note (mutex-based, not sequential block) - Update GitHub Action inputs with new analyzer thresholds

SimplyLiz committed Mar 19, 2026

docs: Add unified code review documentation (ckb review, reviewPR) - New Code-Review.md: checks, formats, policy, health, baselines, compliance - Update Home, Features, User-Guide, MCP-Tools, CI-CD-Integration, Workflow-Examples, Quality-Gates with cross-references Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

SimplyLiz committed Mar 19, 2026

docs: update Home.md for v8.1 release - Update What's New section from v8.0 to v8.1 with refactoring intelligence features - Update preset tool counts to match presets.go (core:21, review:27, refactor:33, docs:27, federation:35, full:92) - Update tool reference count from 80+ to 90+

SimplyLiz committed Feb 1, 2026

docs: re-add planRefactor to refactor preset table Listed with note that it's also in core, since users reading the refactor preset section expect to see refactoring tools there. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Feb 1, 2026

fix: correct preset tool counts to match source code Verified all counts against presets.go: core=21, review=27, refactor=33, federation=35, docs=27, ops=33, full=92 Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Feb 1, 2026

docs: update wiki for v8.1 release - MCP-Tools: Add findCycles, suggestRefactorings, analyzeTestGaps, planRefactor tool docs. Document prepareChange move and extract enhancements with startLine/endLine params. - Presets: Update refactor preset (33 tools), update all preset counts - MCP-Integration: Add v8.1 implementation status, refactoring workflows, update preset table counts Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Feb 1, 2026

docs: document switchProject tool and Cursor CWD workaround - Add switchProject to MCP-Tools.md tool reference and workflows - Update MCP-Integration.md Cursor notes and troubleshooting - Add switchProject to core preset in Presets.md Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 31, 2026

docs: clarify incremental indexing is database-level, not SCIP-level SCIP indexers always run on the full codebase - CKB's "incremental" optimization is selective database import, not per-file SCIP generation. Added clear documentation of this distinction with performance tables. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 28, 2026

docs: update API examples to use index.scip default Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 27, 2026

docs: update default SCIP index path to index.scip - Change default indexPath from .scip/index.scip to index.scip (Sourcegraph standard) - Add documentation for new `ckb config set` and `ckb config get` commands - Update example outputs to reflect new default Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 27, 2026

docs: add Homebrew installation instructions Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 23, 2026

docs: add Grok to supported AI tools across wiki pages Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 22, 2026

docs: update default output format from json to human CLI commands now default to human-readable output instead of JSON. Updated documentation to reflect the new defaults for: - search, refs, impact, arch commands - scan-secrets command - affected-tests command Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 22, 2026

docs: update wiki for v8.0 release - Add compound operations (explore, understand, prepareChange, batchGet, batchSearch) - Add SSE streaming documentation for findReferences and searchSymbols - Add reindex MCP tool section to Index-Management - Add prepareChange tool section to Impact-Analysis - Add Compound Operations section to Prompt-Cookbook - Update all tool counts from 76/77/81 to 80+ Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 20, 2026

docs: update getArchitecture documentation with granularity options - Add granularity parameter (module/directory/file) - Document inferModules and targetPath parameters - Add hard caps for each granularity level - Update API-Reference with query parameters table and examples Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 16, 2026

docs(security): document comprehensive security audit workflow - Add table of 8 configurable scanners with env variables - Document consolidated PR comment feature - Show example PR comment format - Add workflow configuration snippet with all options - Keep simple workflow example for minimal setups Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 15, 2026

docs: add Security page for secret detection feature - New Security.md page documenting scanSecrets MCP tool - Document 26 builtin patterns, scan scopes, and allowlist configuration - Add CI/CD integration examples (pre-commit, GitHub Actions) - Update Home.md with v8.0 security features and links Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

SimplyLiz committed Jan 15, 2026