xg-gh-25/SwarmAI
48.4
Weak · 9 November 2026
362.9k
lines of production code
Python
with TypeScript
5
measurements over time
What this system is
Features
Automated sync script for GitHub Discussions mirror
Added a new Python script (scripts/sync\_discussions.py) that synchronizes the local docs/discussions/ directory with live GitHub Discussions. The tool uses discussion numbers as unique identifiers to prevent duplicates, supports check and write modes, and handles adding new discussions, updating edited content while preserving editorial headers, and rebuilding the README index.
python · high confidence
DOCX skill now supports Amazon Narrative Format and XG style standards
The DOCX skill has been updated with new document style standards. It now includes an 'Amazon Narrative Format' for creating 6-pagers and LT review docs, featuring dense prose, bold lead assertions, and specific appendix structures. Additionally, an 'XG document style standard' has been added, enforcing defaults like Calibri 10.5pt body text, 0.7-inch margins, and KaiTi for Chinese text. The skill also includes comprehensive instructions for using the docx-js library, OOXML schema validation, and redlining workflows.
_backend/skills/s\docx · high confidence
Initial repository structure and security baselines
The repository is initialized with core configuration and documentation files, including a comprehensive .gitignore to prevent leakage of private projects, internal skills, and test artifacts, alongside a .secrets.baseline to track known secret fingerprints and prevent accidental exposure of credentials in future commits.
(repo-wide) · high confidence
Introduce 24/7 backend daemon and channel system with Slack support and security hardening
This change introduces the backend channel subsystem, enabling the agent to connect to external messaging platforms like Slack via a new gateway and adapter architecture. To ensure channels remain active 24/7, a macOS launchd daemon is installed to keep the backend running independently of the desktop app, featuring a guardian watchdog for self-healing restarts and automatic log rotation. The system includes native Slack streaming with reaction controls, per-channel model overrides, and a Slack-native owner approval UI for managing trusted senders. Security is strengthened with an egress redactor that automatically strips credentials and exfiltration URLs from outbound messages, and strict per-channel session isolation with TTL-based eviction.
backend/channels · high confidence
Introduce governed 11-file context system with unified language and security baseline
SwarmAI now ships with a standardized, 11-file context system (backend/context/) that defines the agent's identity, operational rules, security standards, and memory model. This includes AGENT.md for execution directives and mandatory coding gates (R1–R32), SOUL.md for core principles, and a new SECURITY-BASELINE.md enforcing secure-coding rules (A1–B8) on every change. A new CONTEXT.md provides a ubiquitous language glossary to ensure consistent terminology across all 69 skills, while MEMORY.md, EVOLUTION.md, and KNOWLEDGE.md establish a structured, agent-owned memory and correction registry. This replaces ad-hoc context management with a governed, cross-file contract that standardizes how the agent reasons, codes, and remembers.
backend/context · high confidence
Introduce session lifecycle hooks for post-session automation
A new \backend/hooks\ package has been added to execute automated tasks when a session closes, decoupling these operations from the critical chat path. The package includes \WorkspaceAutoCommitHook\ for smart, conventional git commits with code-review gating, \DailyActivityExtractionHook\ for summarizing conversations, \DistillationTriggerHook\ for archiving and curating memory entries, \ContextHealthHook\ for maintaining knowledge indexes, \EvolutionMaintenanceHook\ for pruning stale evolution entries, and \ImprovementWritebackHook\ for capturing lessons into project documentation. Also in this area: Prompt injection hardening and tool-call leak recovery improvements; Artifact lifecycle source classification and project registration fixes.
backend/hooks · high confidence
Introduces new hooks and utilities for audio, streaming, and file-change handling
Adds several new React hooks and utility modules to the desktop frontend: AudioKeepAlive to prevent macOS audio session teardown in WKWebView, useAudioPlayer for Web Audio API playback, and useAttentionQueue for the unified 'Need You' channel. It also introduces streaming-guards and streaming-machine to enforce send-queuing logic and manage the chat lifecycle via a state machine, alongside fileChangedBroker and railSsot to centralize file-change event dispatching and path-matching logic for the Canvas rail. Also in this area: Streamlined chat interface with collapsible tool groups and expert visibility; Improved handling of backend-initiated stream recovery with visual feedback; Stabilize FileViewer close-guard test against race conditions; Artifacts overlay redesign and Canvas drift signals; New evaluation dashboard and onboarding flow with comprehensive test coverage; and 35 more.
desktop · high confidence
Narrative Writing skill introduces guided authoring with templates and quality gates
The narrative-writing skill (v1.3.1) now provides a structured authoring workflow for seven document types—six-pager, PR/FAQ, HLD, LLD, ADR, SOP, and project plan—each with a dedicated template and guided questions. It enforces a five-criteria quality bar, requires reader testing and stakeholder simulation via sub-agents, and mandates weasel-word and AI-ism checks using a provided script and reference lists before delivery.
_backend/skills/s\narrative-writing · high confidence
New 'Repo to DDD' skill generates AI-understandable codebase context
A new skill named 'repo-to-ddd' has been added to the backend, allowing users to transform any codebase into a structured, AI-ready format. The skill produces a \.ai-context/\ directory containing DDD-structured artifacts (PRODUCT.md, TECH.md, IMPROVEMENT.md, PROJECT.md, code-intel.json) and an AGENTS.md entry point, designed to help AI agents navigate, understand, and safely modify the project. It supports multi-language repos, monorepo fan-out, and optional signal ingestion (docs, wikis, Slack) to achieve higher understanding levels. An installer script is included to deploy these artifacts into various IDEs (Claude Code, Cursor, VS Code, etc.).
_backend/skills/s\repo-to-ddd/scripts · high confidence
New Browser Agent skill for DOM-based web automation
Added a new 'Browser Agent' skill that enables DOM-based browser automation using Playwright. This skill allows users to navigate websites, read compressed page content, click elements, fill forms, extract data, and take screenshots. It supports multi-tab workflows, persistent sessions via Chrome DevTools Protocol (CDP), and integrates with MCP tools for simple tasks while falling back to the script for complex interactions. The skill includes a self-contained setup that installs Playwright locally and provides commands for launching, navigating, interacting with elements by index, and managing browser state.
_backend/skills/s\browser-agent · high confidence
New Library skill for mounting and searching external directories
Added the s\_library skill, providing a CLI interface to mount external directories as pointers (without copying data) and search the resulting index. The skill distinguishes between code directories, which are indexed into a symbol graph, and documentation directories, which are chunked and indexed into the shared FTS5 knowledge base at mount time, making them immediately recallable. It supports listing registered mounts and searching across library domains, acting as the agent-facing entry point to the core library mount engines.
_backend/skills/s\library/scripts · high confidence
New PDF processing scripts and Markdown-to-PDF conversion pipeline
The PDF skill now includes a suite of new scripts to handle form data and document conversion. For PDF forms, the system can detect fillable fields, extract field metadata, and populate forms either by updating native PDF form fields or by adding text annotations for non-fillable documents. It also provides utilities to convert PDF pages to images and generate visual validation overlays for bounding boxes. Additionally, a new \md2pdf.sh\ script enables converting Markdown to PDF using either a Tectonic (LaTeX) or WeasyPrint (HTML) engine, supporting professional and minimal styling templates, table of contents generation, and batch processing.
_backend/skills/s\pdf/scripts · high confidence
New PPTX skill for creating, editing, and analyzing PowerPoint presentations
A new PPTX skill has been added to the backend, enabling the system to create, edit, and analyze PowerPoint (.pptx) files. This skill provides a comprehensive set of tools for building presentations from scratch using an HTML-to-PPTX workflow, modifying existing slides, restyling content to match specific themes, and extracting text or generating visual thumbnails. It includes detailed instructions for handling Office Open XML structures, validating documents against ISO-IEC 29500-4 schemas, and managing design elements like typography, color palettes, and layout patterns.
_backend/skills/s\pptx · high confidence
New Remotion-based video composition templates for Pollinate Studio
The Pollinate Studio backend now includes a suite of Remotion templates for generating video content, introducing support for both horizontal (16:9) and vertical (9:16) aspect ratios. These templates provide a visual editing interface via Remotion Studio, allowing users to customize colors, typography, and layout through a defined schema. The package includes specialized compositions for standard videos, short-form vertical content (optimized for platforms like Bilibili), and platform-specific thumbnails (16:9, 4:3, 3:4, 9:16). It also features a library of reusable UI components for rendering data visualizations, code blocks, animated diagrams, and audio waveforms, along with robust error handling and localization support for Chinese and English content.
_backend/skills/s\pollinate/templates/remotion · high confidence
New XLSX skill with formula recalculation and error checking
A new XLSX skill has been added to the backend, enabling the agent to create, edit, and analyze Excel files (.xlsx, .xlsm, .csv, .tsv). The skill provides detailed instructions for financial modeling standards, such as color coding and number formatting, and guides the agent to use Excel formulas rather than hardcoded values to ensure dynamic spreadsheets. It includes a new \recalc.py\ script that leverages LibreOffice to recalculate formulas and detect common Excel errors (e.g., \#DIV/0!, \#REF!), and supports data analysis via DuckDB for SQL-based exploration and pandas for data manipulation.
_backend/skills/s\xlsx · high confidence
New architecture diagrams and design documentation added to assets
The assets directory now includes a comprehensive set of new visual and documentation resources to clarify the system's architecture. A new HTML design document (SwarmAI-Architecture-Design-Doc.html) provides a high-level overview of the agentic OS. Several new SVG diagrams have been added to illustrate specific components: the autonomous pipeline (aidlc-autonomous-pipeline-v4.svg) details the 9-stage, 3-gate workflow; the context engineering diagram (context-engineering.svg) explains the 11-file priority chain and token budget management; the DDD three-layer stack (ddd-three-layer-stack.svg) visualizes the interface, intelligence, and orchestration layers; and the eval architecture (eval-architecture.svg) outlines the decoupled evaluation system. These assets serve as updated reference materials for the platform's engineering and operational structure.
assets · high confidence
New automated code quality, documentation, design review, and security hooks
The \.kiro/hooks\ directory now includes four new user-triggered hooks that automatically analyze and fix code changes. The 'Code Quality Scan & Auto-Fix' hook scans for code smells, anti-patterns, and best-practice violations, automatically applying fixes for High and Medium severity issues. The 'Docs Update Scan & Fix' hook detects when documentation (README, AGENT.md, specs) needs updating based on code changes and applies those updates. The 'PE Design Review, Revise & Propagate' hook reviews newly created design.md files for architectural, security, and correctness issues, revising the design and propagating High/Medium fixes to related requirements and tasks. The 'Security Leak Scan & Auto-Fix' hook scans for hardcoded secrets, credentials, and other security vulnerabilities, automatically fixing Critical severity issues by replacing them with environment variables or config references.
.kiro/hooks · high confidence
New backend utility modules for path resolution, locking, and validation
The backend now includes a new \utils\ package providing shared utilities for core operations. \bundle\_paths\ centralizes logic for locating resource files across development and Tauri production environments, ensuring the daemon can find its configuration and data files regardless of how the binary is deployed. \file\_lock\ introduces cross-platform file locking (supporting both Unix and Windows) to prevent race conditions during concurrent access to shared files. \jsonl\_rotation\ adds automatic rotation for JSONL log files to manage disk usage. Finally, \mcp\_validation\ extracts and standardizes validation logic for MCP configuration entries, including security checks to prevent environment variables from pointing to the system database. Also in this area: Introduce channel permission tiers and sender identity model.
backend/utils · high confidence
New frontend-design skill with BM25 search and design system generation
The \backend/skills/s\_frontend-design/scripts\ area now contains the core logic for a new frontend-design skill. This includes \core.py\, which implements a BM25 search engine over local CSV data for UI/UX style guides, colors, charts, and other design domains; \design\_system.py\, which aggregates these search results to generate comprehensive design system recommendations; and \search.py\, a CLI entry point that exposes these capabilities via command-line arguments, including options for domain-specific search, stack-specific guidelines, and persistent design system output generation.
_backend/skills/s\frontend-design/scripts · high confidence
New modular TTS engine with multi-backend support and SSML enhancements
The Pollinate skill now includes a dedicated, self-contained Text-to-Speech (TTS) module that supports multiple synthesis backends (Azure, Amazon Polly, Edge, CosyVoice, Doubao, ElevenLabs, OpenAI, and Google). Users can select their preferred provider via environment variables or user preferences, with Edge TTS available as a free, keyless default. The engine features advanced SSML processing for improved pronunciation, including Chinese polyphone disambiguation, abbreviation expansion, and automatic language switching for English terms within Chinese narration. It also provides precise word-boundary tracking for accurate subtitle (SRT) generation and section timing synchronization for video production.
_backend/skills/s\pollinate/scripts/tts · high confidence
New signal feed adapters and secure HTTP client infrastructure
This change introduces a new signal feed adapter system in \backend/jobs/adapters\ to populate the product's signal pipeline with diverse, real-time data. New adapters include \eastmoney\_market\ for Chinese A-share market movers, \trending\ for Chinese social media hot searches (Weibo, Zhihu, etc.), \weibo\_trending\ for keyword-matched Weibo posts, \github\_trending\ for trending repositories, \github\_releases\ for tracked project updates, \github\_community\ for scanning specific repos for engagement opportunities, \hacker\_news\ for AI-filtered stories, and \rss\ for standard feed parsing. To support these external fetches securely, a shared \http\_client\ module is added, featuring an SSRF egress guard that validates resolved IPs to block private, loopback, and metadata addresses, and enforces HTTPS-only connections. The implementation also includes retry logic for transient network errors and disables environment proxy variables to ensure predictable, thread-safe outbound traffic.
backend/jobs/adapters · high confidence
New static audit script for CLI argument documentation drift
A new Python script has been added to the backend audit tools to detect mismatches between command-line arguments defined in skill scripts and those documented in INSTRUCTIONS or SKILL markdown files. This tool scans skills for argparse flags and reports which ones are missing from the documentation, helping maintain consistency between code and user-facing docs.
backend/scripts/audits · high confidence
Pollinate content production engine scripts
Added a suite of Python scripts in backend/skills/s\_pollinate/scripts to support the Pollinate content production engine. These include brand\_chart.py for applying direction tokens to Excel charts, check\_prereqs.py for validating environment dependencies, check\_rpv.py and check\_specs.py for automated quality and platform specification validation of video assets, convergence\_gate.py for enforcing poster HTML quality layers, cross\_format\_check.py for verifying consistency across multi-track outputs, deck\_notes\_injector.py for processing PowerPoint speaker notes, font\_link\_backfill.py for managing Google Fonts CDN links, and format\_recommend.py for recommending production tracks based on audience and context.
_backend/skills/s\pollinate/scripts · high confidence
Slack channel adapter with dual connectivity paths
Introduces a new Slack channel adapter that supports both Socket Mode (WebSocket) and HTTP polling fallback. This allows the system to connect to Slack without requiring a public webhook URL, and automatically switches to polling if the WebSocket connection is blocked by corporate proxies or VPNs. The adapter also includes logic for handling Slack API limits, authentication errors, and message formatting.
backend/channels/adapters · high confidence
Security
Adds application-layer authentication for Hive deployments
Introduces a new pure-ASGI authentication middleware (HiveAuthMiddleware) that enforces HTTP Basic authentication on all API routes in Hive mode, providing a defense-in-depth layer alongside the existing Caddy edge gate. This change ensures that privileged endpoints (such as workspace and job management) require valid credentials even if the network edge is bypassed, while remaining a transparent pass-through in non-Hive modes to avoid locking out local users. The implementation is designed to be fail-closed and compatible with streaming responses, preventing potential security risks identified by CloudSec.
backend/middleware · high confidence
Behavioural changes
1 commit (0 fixes) modifying (repo-wide)
A change to existing behaviour in (repo-wide) — 1 commit, 55 files: “feat: add external App canaries”.
(repo-wide) · medium confidence · unverified
1 commit (0 fixes) modifying backend/skills/s\_pollinate/templates/html-report
A change to existing behaviour in backend/skills/s\_pollinate/templates/html-report — 1 commit, 4 files: “skills: delete 30 built-in skills nobody used”.
_backend/skills/s\pollinate/templates/html-report · medium confidence · unverified
1017 commits (354 fixes) modifying swarmai-backend
A change to existing behaviour in swarmai-backend — 1017 commits (354 fixs), 135 files: “feat: zero-touch Brain-to-Hive — no step from New Brain to a finished Hive run waits for the person”; “fix(stability): batch 2 — per-call gates get their own pool; one incremental run index”; “code graph: a graph an older parser wrote rebuilds itself”; and 1009 more (refs \#2, \#3, \#112, \#114, \#1, …).
swarmai-backend · medium confidence · unverified
11 commits (1 fix) modifying backend/templates/eval/fixtures/workspace
A change to existing behaviour in backend/templates/eval/fixtures/workspace — 11 commits (1 fix), 39 files: “test(eval): the New Brain library case carries the two error classes seen on a real setup”; “feat(brain): the entry contract carries where each section's content comes from and what to read whole”; “feat(cdf): a claim without exceptions rests on the code, not a description”; and 8 more.
backend/templates/eval/fixtures/workspace · medium confidence · unverified
143 commits (41 fixes) modifying backend/core/run\_protocol
A change to existing behaviour in backend/core/run\_protocol — 143 commits (41 fixs), 31 files: “fix(backend): log the three silent excepts that kept the baseline red”; “next says early when main's commits touch the run's files”; “B-e: three-way re-audit fixes across the CLI/daemon boundary”; and 140 more.
_backend/core/run\protocol · medium confidence · unverified
160 commits (57 fixes) modifying backend/jobs/handlers
A change to existing behaviour in backend/jobs/handlers — 160 commits (57 fixs), 18 files: “feat: zero-touch Brain-to-Hive — no step from New Brain to a finished Hive run waits for the person”; “code graph: a graph an older parser wrote rebuilds itself”; “feat(eval): rename eval-nightly to eval-scheduled + score-drift hard gate (run\_95d9acbc)”; and 157 more.
backend/jobs/handlers · medium confidence · unverified
1645 commits (670 fixes) modifying backend/core
A change to existing behaviour in backend/core — 1645 commits (670 fixs), 224 files: “S4: hook chain keeps every hook's hookSpecificOutput.additionalContext”; “Stop test boots killing the daemon's CLIs; retry dead continuations”; “fix(agent): Exclude None-valued context\_token\_budget from build\_agent\_config”; and 1634 more (refs \#14, \#5, \#2, \#112, \#114, …).
backend/core · medium confidence · unverified
18 commits (3 fixes) modifying backend/skills/s\_frontend-design/data
A change to existing behaviour in backend/skills/s\_frontend-design/data — 18 commits (3 fixs), 48 files: “前端 chat 回合展示按 mock v7.11 落地(只做前端 + design-judgment 文档,后端数据另起 run)。规格 = K”; “Chat turn first-level structure: direct runs are not pipelines; no”; “docs(skill): budget the space before touching CSS + measurement pitfalls”; and 15 more.
_backend/skills/s\frontend-design/data · medium confidence · unverified
19 commits (5 fixes) modifying backend/core/brain\_bootstrap
A change to existing behaviour in backend/core/brain\_bootstrap — 19 commits (5 fixs), 8 files: “feat: zero-touch Brain-to-Hive — no step from New Brain to a finished Hive run waits for the person”; “new brain: one entry contract, generated from the rules bootstrap-apply enforces (r12 layer 1)”; “feat(brain): the entry contract carries where each section's content comes from and what to read whole”; and 16 more.
_backend/core/brain\bootstrap · medium confidence · unverified
2 commits (0 fixes) modifying (repo-wide)
A change to existing behaviour in (repo-wide) — 2 commits, 29 files: “initial commit”; “delete no used files”.
(repo-wide) · medium confidence · unverified
2 commits (0 fixes) modifying backend/schemas/\_\_pycache\_\_
A change to existing behaviour in backend/schemas/\_\pycache\\_ — 2 commits, 11 files: “initial commit”; “remove cache files”.
backend/schemas/\\pycache\\_ · medium confidence · unverified_
2 commits (0 fixes) modifying backend/templates/eval/fixtures/workspace/Projects/Demo
A change to existing behaviour in backend/templates/eval/fixtures/workspace/Projects/Demo — 2 commits, 5 files: “eval: the two new offline cases test what their rule says, on a fixture that makes the right answer possible”; “eval: 8 offline product cases run in a shipped fixture workspace (design §3.8)”.
backend/templates/eval/fixtures/workspace/Projects/Demo · medium confidence · unverified
29 commits (4 fixes) modifying backend/skills/s\_project-manager/scripts
A change to existing behaviour in backend/skills/s\_project-manager/scripts — 29 commits (4 fixs), 2 files: “feat: zero-touch Brain-to-Hive — no step from New Brain to a finished Hive run waits for the person”; “refactor(project-agent): one standard definition per Brain, no confirmation”; “ddd: one skeleton for every brain, entry taint + one renderer, P6 checks the skeleton”; and 26 more.
_backend/skills/s\project-manager/scripts · medium confidence · unverified
3 commits (0 fixes) modifying backend/bin
A change to existing behaviour in backend/bin — 3 commits, 1 file: “S3a: the backend program runs Python itself (py / pipeline-v1 subcommands)”; “dev shim: stop when backend/.venv is missing instead of using PATH python3”; “cdf v2 stage B + cutover work (squashed from kiro/cdf-flow-backup for the rebase onto main)”.
backend/bin · medium confidence · unverified
35 commits (17 fixes) modifying hive
A change to existing behaviour in hive — 35 commits (17 fixs), 11 files: “fix: Hive update flow hardening — 5 structural gaps (H1-H5)”; “feat: unified release pipeline — prod.sh release-all + Hive verification + CI/CD”; “fix: 7 PE findings — hook signature, orphan edges, resource lifecycle”; and 31 more.
hive · medium confidence · unverified
4 commits (0 fixes) modifying assets/readme
A change to existing behaviour in assets/readme — 4 commits, 28 files: “docs: Core Logic page in the Autonomous Guideline, drawn as a loop”; “docs(readme): write the vision in full: verified delivery in the north star, Brain versions for parallel Hives, the one-line positioning”; “docs: Autonomous Guideline in the README, with the full deck in docs/decks”; and 1 more.
assets/readme · medium confidence · unverified
4 commits (0 fixes) modifying backend/skills/s\_ddd-distribute/scripts
A change to existing behaviour in backend/skills/s\_ddd-distribute/scripts — 4 commits, 1 file: “把 SecDLC COE-\>Pattern-\>Scan 扫描引擎从全词法检索升级为语义优先/词法兜底(S1,S2,S3,S5 转 semanti”; “feat(ddd-distribute): --with-enablement ships engine to bare foreign hosts”; “P7: product skills stop assuming a SwarmAI source checkout”; and 1 more.
_backend/skills/s\ddd-distribute/scripts · medium confidence · unverified
4 commits (1 fix) modifying backend/templates/ddd-skills/s\_ddd-persist/scripts
A change to existing behaviour in backend/templates/ddd-skills/s\_ddd-persist/scripts — 4 commits (1 fix), 1 file: “memory entry spans: change MEMORY/EVOLUTION whole entries at a time”; “feat(ddd): restore sedimented philosophy in s\_ddd-persist + s\_ddd-manager”; “cdf v2 stage B + cutover work (squashed from kiro/cdf-flow-backup for the rebase onto main)”; and 1 more.
_backend/templates/ddd-skills/s\ddd-persist/scripts · medium confidence · unverified
440 commits (148 fixes) modifying backend/routers
A change to existing behaviour in backend/routers — 440 commits (148 fixs), 36 files: “feat: zero-touch Brain-to-Hive — no step from New Brain to a finished Hive run waits for the person”; “fix(stability): batch 2 — per-call gates get their own pool; one incremental run index”; “Artifact subsystem P0: single trust boundary for all caller-supplied path identifiers”; and 436 more (refs \#4, \#112, \#114, \#9).
backend/routers · medium confidence · unverified
5 commits (1 fix) modifying backend/skills/s\_context-hygiene/scripts
A change to existing behaviour in backend/skills/s\_context-hygiene/scripts — 5 commits (1 fix), 1 file: “Remove in-prompt markdown index generation: delete refresh\_knowledge\_ind”; “feat(skills): add s\_context-hygiene — context-file cleanup methodology + read-only scanner”; “fix(skills): stop publishing identifiers only the author can resolve”; and 2 more.
_backend/skills/s\context-hygiene/scripts · medium confidence · unverified
59 commits (20 fixes) modifying backend/core/code\_intel
A change to existing behaviour in backend/core/code\_intel — 59 commits (20 fixs), 8 files: “feat: zero-touch Brain-to-Hive — no step from New Brain to a finished Hive run waits for the person”; “fix: 7 PE findings — hook signature, orphan edges, resource lifecycle”; “feat(code-intel): language-agnostic bounded-domain clustering (graph\_clusters)”; and 55 more.
_backend/core/code\intel · medium confidence · unverified
595 commits (212 fixes) modifying backend/scripts
A change to existing behaviour in backend/scripts — 595 commits (212 fixs), 91 files: “run-commit 的 deliverable-B copy(REPORT.md 进 .aidlc/deliverables/\<run\_id\>”; “feat: zero-touch Brain-to-Hive — no step from New Brain to a finished Hive run waits for the person”; “Add a non-gate-eligible eval tier (e.g. 'live'/'integration') for determ”; and 592 more (refs \#5, \#4, \#8, \#9, \#3).
backend/scripts · medium confidence · unverified
6 commits (1 fix) modifying backend/skills/s\_autonomous-pipeline/scripts
A change to existing behaviour in backend/skills/s\_autonomous-pipeline/scripts — 6 commits (1 fix), 2 files: “feat(pipeline): Group Gate-2 fixes by cause before any fix (FIXING\_SPEC)”; “让 cultivation 的校准输入不再是死的,并让「全被丢弃」在最外层可见”; “Fix the thing”; and 3 more.
_backend/skills/s\autonomous-pipeline/scripts · medium confidence · unverified
6 commits (2 fixes) modifying backend/skills/s\_radar-todo/scripts
A change to existing behaviour in backend/skills/s\_radar-todo/scripts — 6 commits (2 fixs), 1 file: “chore: clean up legacy specs, unused pages, and stale components”; “refactor: centralize hardcoded Path.home()/.swarm-ai to config.get\_app\_data\_dir()”; “tests on main green again; skill scripts resolve the app data dir one way”; and 3 more.
_backend/skills/s\radar-todo/scripts · medium confidence · unverified
64 commits (13 fixes) modifying backend/core/ingestion/sources
A change to existing behaviour in backend/core/ingestion/sources — 64 commits (13 fixs), 9 files: “fix(cdf): a background source waits for a busy store instead of losing the unit”; “CDF: a gap closes when an event says it is done, whichever session opened it”; “cdf review: a long session's fork that cannot answer falls back to the record; relations asked again fresh”; and 61 more.
backend/core/ingestion/sources · medium confidence · unverified
7 commits (1 fix) modifying backend/generated
A change to existing behaviour in backend/generated — 7 commits (1 fix), 5 files: “apps index: regenerate after the Pollinate script edits”; “fix: regenerate the App index; select its freshness test by App source”; “apps index: regenerate after the frontend-design skill edit (knowledge addressing run 3)”; and 4 more.
backend/generated · medium confidence · unverified
86 commits (14 fixes) modifying backend/core/ingestion
A change to existing behaviour in backend/core/ingestion — 86 commits (14 fixs), 20 files: “cdf session review: metrics in Flow health (§12)”; “fix(cdf): a background source waits for a busy store instead of losing the unit”; “projects: a name in another case gets the existing folder's spelling”; and 83 more.
backend/core/ingestion · medium confidence · unverified
9 commits (3 fixes) modifying backend/skills/s\_notify
A change to existing behaviour in backend/skills/s\_notify — 9 commits (3 fixs), 3 files: “skills: delete 30 built-in skills nobody used”; “fix: Voice Conversation Mode E2E fixes + security hardening (17 findings)”; “S3c: skills run Python with the backend program's own interpreter”; and 6 more.
_backend/skills/s\notify · medium confidence · unverified
98 commits (40 fixes) modifying scripts
A change to existing behaviour in scripts — 98 commits (40 fixs), 20 files: “fix(deploy): Isolate the build probe, drain turns before restart”; “feat(docs): document metadata convention — created/updated dates + lint + auto-update”; “fix: structural prevention for skill name case mismatch”; and 95 more.
scripts · medium confidence · unverified
Backend schema overhaul: new capabilities, filesystem-backed configs, and agent model changes
This change introduces several new Pydantic schemas for previously missing capabilities: autonomous jobs (with status and scheduling fields), channel management (including per-channel model overrides and session tracking), and context assembly previews (with layer details and thread binding). It replaces the database-backed MCP configuration schemas with a two-layer file-system-based system (catalog and dev entries) and simplifies the settings schema by removing the detailed AWS/Anthropic credential models. Agent schemas are updated to increase the maximum turn limit to 2000, rename 'skill\_ids' to 'allowed\_skills', deprecate per-agent sandbox configuration in favor of app-level settings, and add system agent flags. Additionally, skill schemas are simplified to be filesystem-based, removing version control and complex metadata in favor of folder-name identity and tiered loading.
backend/schemas · high confidence
Backend test infrastructure overhaul and model registry centralization
The backend test suite is now memory-safe, splitting execution into batches via a new safe\_test\_runner.py to prevent 9GB+ RSS spikes, with Hypothesis profiles standardized in conftest.py. Model configuration is centralized in a new model\_registry.py, replacing five drifted hardcoded copies with a single authority that makes claude-opus-4-8 the default flagship. Legacy AWS Bedrock tool-use demos and the old .env configuration have been removed, and the .env.example has been updated to reflect the new desktop-focused SQLite and Tauri settings. Also in this area: Fixes model glitches and adds Opus 5.5 support; Prompt injection hardening and tool-call leak recovery improvements.
backend · high confidence
Removal of DynamoDB support and migration to SQLite-only architecture
The backend database layer has been simplified to use SQLite exclusively, removing all DynamoDB integration code and configuration. The \dynamodb.py\ client module has been deleted, and the database initialization logic in \\_\init\\_.py\ no longer checks for a \DATABASE\_TYPE\ environment variable to switch between providers; it now always instantiates the \SQLiteDatabase\. Consequently, the abstract base class \BaseDatabase\ has been updated to remove the \skills\ and \skill\_versions\ table properties, while adding new properties for channel-related tables (\channels\, \channel\_sessions\, \channel\_messages\, \channel\_user\_identities\). This change eliminates the need for AWS credentials and cloud infrastructure for database storage, consolidating all data persistence to the local SQLite engine.
backend/database · high confidence
Fixes
3 commits (2 fixes) fixing .claude
A fix in .claude — 3 commits (2 fixs), 2 files: “feat: PreToolUse hook to sanitize pytest commands — prevent session eviction”; “fix(hooks): structural prevention for C012 — block WebFetch on SPA domains”; “Review fixes: cap every pytest run, trust no bad timeout, stop rewriting commands”.
.claude · medium confidence · unverified
41 commits (24 fixes) fixing backend/hive
A fix in backend/hive — 41 commits (24 fixs), 5 files: “fix: Hive update flow hardening — 5 structural gaps (H1-H5)”; “fix: 7 PE findings — hook signature, orphan edges, resource lifecycle”; “fix: remove CloudFront /assets/\* cache behavior — auth header was stripped”; and 37 more.
backend/hive · medium confidence · unverified
7 commits (4 fixes) fixing backend/skills/s\_pollinate/templates/html-deck/shared
A fix in backend/skills/s\_pollinate/templates/html-deck/shared — 7 commits (4 fixs), 5 files: “feat(ddd-distribute): --with-enablement ships engine to bare foreign hosts”; “fix(skills): stop publishing identifiers only the author can resolve”; “fix(security): Bind PDF export server to loopback”; and 4 more.
_backend/skills/s\pollinate/templates/html-deck/shared · medium confidence · unverified
90 commits (50 fixes) fixing desktop/src-tauri
A fix in desktop/src-tauri — 90 commits (50 fixs), 6 files: “fix(desktop): Add graceful shutdown before force-kill on app close”; “fix: Code quality and security fixes from scan review”; “fix(daemon): self-healing restart + lsof elimination + rollback hardening”; and 87 more (refs \#1, \#112, \#114).
desktop/src-tauri · medium confidence · unverified
Job system moved to product-level package with in-process scheduler and unified model resolution
The background job infrastructure has been consolidated into a new \backend/jobs\ package, replacing the previous workspace-level layout. The scheduler now runs in-process within the daemon's asyncio loop, eliminating the need for external \launchd\ plists and their associated authentication failures. A new \model\_resolve\ module serves as the single source of truth for job model selection, ensuring all headless tasks use the correct flagship model with the 1M context window suffix, which fixes the recurring '400 Input is too long' errors. Additionally, the system introduces a robust credential resolution strategy for AWS Bedrock that pre-resolves tokens to handle launchd contexts, and adds a \channel\_monitor\_fallback\ to maintain Slack monitoring via bot tokens when the primary MCP-based auth expires.
backend/jobs · high confidence
Token estimation now uses the canonical calibrated model instead of a flawed shell heuristic
The \s\_estimate-tokens\ skill has replaced its previous bash-based estimation logic (which relied on \wc -w\ and a hardcoded 200K token window) with a Python script that delegates to the canonical \ContextDirectoryLoader.estimate\_tokens\. This change ensures that token counts for CJK and other languages match the actual tokenizer used by the prompt assembly, eliminating significant under-counting errors. The script dynamically discovers the repository root to import the single source of truth for the estimator and uses the effective 1M-context budget (derived as 141K tokens) for percentage calculations, failing loudly if the canonical module cannot be located rather than silently falling back to incorrect heuristics.
_backend/skills/s\estimate-tokens/scripts · high confidence
Test coverage
2154 commits adding/updating tests in backend/tests; Added comprehensive test coverage for the Code Intelligence platform; Added test coverage for legacy code cleanup safety invariants; Added tests for HTML deck track and style selection logic; Added tests for keep-mounted per-tab chat views and cross-tab isolation; Added validation tests for frontend-design slide assets and previews.
Dependencies
Rebrand to SwarmAI and upgrade core dependencies
The application has been rebranded from 'Claude Agent Platform' to 'SwarmAI' across the backend, desktop, and root manifests. This release upgrades the backend to use \claude-agent-sdk\ version 0.2.109 and introduces new capabilities including Slack integration (\slack-bolt\), voice transcription (\amazon-transcribe\), and code intelligence tools (\tree-sitter\, \watchfiles\). The desktop application now supports automatic updates via the Tauri updater plugin, features a unified file viewer with PDF support, and includes a new integrated terminal using \portable-pty\. Additionally, the Python test suite now defaults to serial execution to prevent deadlocks, and the backend version has been bumped to 2.1.0.
(dependencies) · high confidence
Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.
How this codebase got here
Score
- CAI 50 → 48 (-1.4)
- Rubric changed (rubric-2026.10.5 → rubric-2026.11.2) — scores are not directly comparable.
Lenses
- Code Health 56 → 59 (+3.1)
- Architecture 87 → 68 (-19.5)
- Maturity 52 → 80 (+27.9)
- Readiness 49 → 41 (-8.8)
- Security 76 → 75 (-0.7)
- Accessibility 45 → 46 (+0.9)
- Performance 60 → 60 (+0.0)
Resolved (852)
- ActivityFeed.extractActivities (cognitive 39) (desktop/src/pages/chat/components/ActivityFeed.tsx)
- ActivityFeed.extractActivities (cyclomatic 26) (desktop/src/pages/chat/components/ActivityFeed.tsx)
- AuthConfigPanel.AuthConfigPanel::handleVerify (cognitive 19) (desktop/src/components/settings/AuthConfigPanel.tsx)
- AuthConfigPanel.AuthConfigPanel::handleVerify (cyclomatic 16) (desktop/src/components/settings/AuthConfigPanel.tsx)
- BrainHub.ReviewView (cognitive 25) (desktop/src/components/layout/BrainHub.tsx)
- BrainHub.ReviewView (cyclomatic 24) (desktop/src/components/layout/BrainHub.tsx)
- BrainHub.hunkSummary (cognitive 16) (desktop/src/components/layout/BrainHub.tsx)
- CMBrainOverlay.ContextTab (cyclomatic 16) (desktop/src/components/layout/CMBrainOverlay.tsx)
- CMBrainOverlay.KnowledgeGraph (cognitive 19) (desktop/src/components/layout/CMBrainOverlay.tsx)
- CMBrainOverlay.KnowledgeGraph (cyclomatic 16) (desktop/src/components/layout/CMBrainOverlay.tsx)
- CMBrainOverlay.ProposalCard (cyclomatic 16) (desktop/src/components/layout/CMBrainOverlay.tsx)
- CanvasOutputRail.CanvasOutputRail (cognitive 24) (desktop/src/components/file-viewer/CanvasOutputRail.tsx)
- Change coupling: useCanvasAutoSurface.ts ↔ useReferencedFiles.ts (desktop/src/hooks/useCanvasAutoSurface.ts)
- ChapterProgressBar.ChapterProgressBar (cognitive 16) (backend/skills/s_pollinate/templates/remotion/components/ChapterProgressBar.tsx)
- ChatPage.ChatPage::handleAnswerQuestion (cognitive 19) (desktop/src/pages/ChatPage.tsx)
- ChatPage.ChatPage::handlePermissionDecision (cognitive 19) (desktop/src/pages/ChatPage.tsx)
- ChatPage.ChatPage::handlePermissionDecision (cyclomatic 16) (desktop/src/pages/ChatPage.tsx)
- ChatPage.ChatPage::handleStop (cognitive 26) (desktop/src/pages/ChatPage.tsx)
- ChatPage.ChatPage::handleStop (cyclomatic 19) (desktop/src/pages/ChatPage.tsx)
- ClassTooLong: AuthConfigPanel (desktop/src/components/settings/AuthConfigPanel.tsx)
- …and 832 more
New (1793)
- AppConfigManager.set_secret (cognitive 18) (backend/core/app_config_manager.py)
- AssistantMessageView.BackgroundFinishedCard (cognitive 17) (desktop/src/pages/chat/components/AssistantMessageView.tsx)
- AttachmentPreviewCard.AttachmentPreviewCard (cyclomatic 16) (desktop/src/components/chat/AttachmentPreviewCard.tsx)
- AuditStore.append_batch (cognitive 17) (backend/core/run_protocol/store.py)
- AuditStore.copy_to_session (cognitive 16) (backend/core/run_protocol/store.py)
- AwsAccountTab.AwsAccountTab (cognitive 81) (desktop/src/components/settings/AwsAccountTab.tsx)
- AwsAccountTab.AwsAccountTab (cyclomatic 88) (desktop/src/components/settings/AwsAccountTab.tsx)
- BrainHub.BrainBrowse (cyclomatic 18) (desktop/src/components/layout/BrainHub.tsx)
- BrainHub.BrainHeader (cyclomatic 17) (desktop/src/components/layout/BrainHub.tsx)
- BrainHubHome.BrainHubHome (cognitive 18) (desktop/src/components/layout/BrainHubHome.tsx)
- BrainHubHome.BrainHubHome (cyclomatic 25) (desktop/src/components/layout/BrainHubHome.tsx)
- BrainLearnedView.BrainLearnedView (cognitive 17) (desktop/src/components/layout/BrainLearnedView.tsx)
- BrainLearnedView.BrainLearnedView (cyclomatic 18) (desktop/src/components/layout/BrainLearnedView.tsx)
- BrainLearnedView.Entry (cognitive 24) (desktop/src/components/layout/BrainLearnedView.tsx)
- BrainLearnedView.Entry (cyclomatic 25) (desktop/src/components/layout/BrainLearnedView.tsx)
- BrainMap.BrainMap (cognitive 16) (desktop/src/components/layout/BrainMap.tsx)
- BrainMap.BrainMap (cyclomatic 16) (desktop/src/components/layout/BrainMap.tsx)
- BrainMapView.EvalPanel (cognitive 21) (desktop/src/components/layout/BrainMapView.tsx)
- BrainMapView.EvalPanel (cyclomatic 23) (desktop/src/components/layout/BrainMapView.tsx)
- Bridge._valid (cognitive 33) (backend/core/swarm_out_bridge.py)
- …and 1773 more
Changes since last survey
- 300 commits — 215 feature/other, 85 fixes
By area
- backend/core — 112 commits
- docs/design — 45 commits
- desktop/src — 35 commits
- backend/tests — 31 commits
- (root) — 13 commits
- backend/skills — 8 commits
- .github/workflows — 7 commits
- backend/scripts — 7 commits
- docs/releases — 7 commits
- backend/templates — 5 commits
- desktop/src-tauri — 5 commits
- docs/discussions — 5 commits
- backend/context — 4 commits
- assets/readme — 3 commits
- backend/jobs — 3 commits
- (repo) — 1 commit
- backend/channels — 1 commit
- backend/desktop_main.py — 1 commit
- backend/experts — 1 commit
- backend/generated — 1 commit
Notable commits
- fix: Canvas: no fixed 900px drag ceiling; on a wide screen it goes as wide as chat allows
- fix: Merge origin/main: take the revert of PRs #112-#114 into history
- fix: Revert "fix(task_checks): a task's setup gets the python setup-python gives a CI job"
- fix: Review fixes for the delivery-friction change (correctness + security, one pass)
- fix: Sign-in errors: right words per sign-in, "not connected" said as it is, a fix button for every path
- fix: Swarm Out: six places a run got stuck with no way out, fixed
- fix: design: knowledge reaches judgment r4 — keep only fixes for problems observed live
- fix: fix(audit): search everything that reaches the agent; count first-stated decisions apart
- fix: fix(audit): the mistake audit reads what the brain held, not one narrow guess
- fix: fix(audit): the monthly mistake audit reads each mistake and entry whole
- fix: fix(backend): Make the backend CI steps green (BVT ratchets, fresh-install probe, lints)
- fix: fix(bedrock): a reply the provider's content filter withheld is 'filtered', not a missing tool call
- fix: fix(brain): a new Brain has no dangling reference; the untouched persona stub does not package
- fix: fix(brain): a package.json script's runner is found on disk, so the source check agrees with the draft
- fix: fix(brain): a redraft folds a twin check an earlier Project Agent draft kept
- fix: fix(brain): a root README with no extension is read as the repository's README
- fix: fix(brain): fix hot spots are code files; tests are tagged, bookkeeping files dropped
- fix: fix(brain): one check declared in two spellings is one Project Agent validation
- fix: fix(brain): package scripts run with the repo's package manager; a global install is not setup
- fix: fix(brain): relocate moves the mirror as it is — moving deletes nothing
- …and 280 more
Architecture
- Unchanged — 0 containers · 2 contexts · 0 edges
Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.
Survey your own repository
xg-gh-25/SwarmAI was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.
About this page
- The score is its most recent published measurement, taken on 9 November 2026 at a pinned commit. It is not a live figure and does not change until the project is measured again.
- Measured at commit 130052ef077bcc717ffbe21546399eeb37b26225 — the exact code this score is about.
- Scored under rubric-2026.11.2 — the same rubric and the same method as every other entry in this index.
- Measured by Canine Code Analyzer (cca.canine.dev) using cca measure-2026.11.3+preprod-bb0cb800cd0f.