ECC 2.2.1

ECC 2.2.1 is a bug and security patch for ECC 2.2. It keeps the published
v2.2.0 history immutable. These notes describe the prepared patch; publication
and signing evidence are tracked separately in the release checklist.

Security and data protection

• GateGuard and governance capture recognize destructive PowerShell commands,
  including the native PowerShell tool path. Dynamic command handling prevents
  later variable assignments from concealing earlier unresolved invocations
  (#2961).
• Relative GateGuard exemption globs stay within the project root. Explicit
  absolute exemptions remain supported
  (#2921).
• Installer writes reject collisions with untracked user-owned files. Failed
  installs refresh ownership hashes only for files they actually wrote, preserving the previous
  ownership hashes of untouched managed files
  (#2964).
• Guided setup revalidates its preview before ownership filtering, so files
  appearing between preview and apply cause a clear retry instead of a false
  success. Existing identical user files stay outside ECC ownership.
• Uninstall respects ECC_DRY_RUN=1, including legacy Codex paths, and rejects
  invalid dry-run values instead of silently allowing deletion
  (#2952).
• Observer analysis retains observations on unsuccessful or unconfirmed
  processing. Exit code zero alone no longer permits archival
  (#2971).
• The Yarn lockfile updates toml to 4.3.0, matching the npm lockfile and
  removing the affected older resolution.

Hooks and installation

• Manual Claude installs register ECC-owned hook entries in Claude settings.
  Repair, consent changes, and uninstall reconcile those entries while
  preserving unrelated settings. Atomic settings updates check directory
  identity and retry detected concurrent edits
  (#2992).
• Direct hook entrypoints handle larger JSON payloads with bounded, UTF-8-safe
  reads instead of silently truncating valid inputs. Existing production
  wrapper limits remain unchanged
  (#2924).
• The Pi adapter selects an actual Node runtime instead of recursively
  executing a compiled OMP host as Node
  (#2909).
• Installer listing and control-pane help avoid eager third-party dependency
  loading. Features that require absent runtime packages report the missing
  dependency explicitly
  (#2994).
• Autonomous harness setup documentation replaces nonexistent package names
  and unsupported CLI flags with documented interfaces, and distinguishes
  session scheduling from a durable external scheduler
  (#2957).

Installer and release-surface hardening

• Public and packaged install docs now consistently point at the published
  ecc-universal commands instead of stale or unrelated package names.
• The AdaL adapter docs use the correct npx ecc-universal doctor --target adal
  command.
• Claude setup preflights git before provider-specific work starts, so missing
  prerequisites fail fast with the right action.
• Guided setup dry runs use isolated HOME, config, XDG, temp, and Windows app
  data roots to avoid ambient host state affecting review or tests.
• The exact packed artifact now has stronger lifecycle coverage for Claude and
  Kimi setup, update, doctor, repeat install, uninstall, and dry-run flows.
• Identifier regression coverage blocks stale ecc, ecc-install, and other
  mismatched release-path commands from creeping back into user-facing docs.

Current-main documentation included in this patch

• The canonical Itô workflow now documents ecc ito accept <ticket-id> and the
  ito_accept MCP tool.
• Acceptance is explicitly bounded to buyer-authority routing. It routes the
  active desk quote to human review and does not claim to place a trade.

Provenance boundary

• v2.2.1 is intended to be a signed annotated tag on exact green main.
• v2.2.0 remains the immutable historical unsigned exception. Do not move,
  recreate, or reuse that tag.

Scope and limitations

• Plugin dependency handling does not bundle or automatically install missing
  modules. Database and schema-validation features require their declared
  runtime dependencies.
• Ownership protection covers untracked collisions and failed-install
  checkpoints. Successful upgrades retain the existing contract for replacing
  previously managed files. Back up intentional edits before upgrading.
• This patch does not introduce new harness platforms or claim that every
  open community issue is resolved.

Upgrade

After the release workflow publishes 2.2.1 and verifies registry integrity,
install or update the package, then run the same ECC command path you already
use. Until publication completes, the exact-version command below returns E404.

npm install -g ecc-universal@2.2.1
ecc doctor

For first-time or guided terminal setup:

npx ecc-universal setup

The native Claude marketplace path remains supported:

/plugin marketplace add https://github.com/affaan-m/ECC
/plugin install ecc@ecc

ECC 2.2 is here, led by one guided installer for Claude Code, Codex, and Kimi.
It also adds native Antigravity 2.0 support and an experimental, opt-in Nasiko
CLI lifecycle bridge.

The release pipeline is stricter too: the exact npm archive must pass lifecycle
tests on Linux, macOS, and Windows before latest moves.

One guided path into ECC

Start the guided installer with:

npx --yes ecc-universal@2.2.0 install --guided

The installer checks every selected harness before writing files, then applies
the setup as one operation. Provider-owned model and authentication settings
stay under the provider's control. The terminal welcome appears only after a
successful interactive install.

Nasiko: experimental, opt-in, and explicit

ECC 2.2 can inspect, install, and remove one pinned Nasiko v0.1.0 CLI build:

npm install -g ecc-universal@2.2.0
ecc nasiko status --json
ecc nasiko install --version v0.1.0 --dry-run --json
ecc nasiko install --version v0.1.0 --yes --json

Downloads are content-addressed and SHA-256 verified, status is read-only, and
uninstall protects modified or unqualified binaries. The dashboard above is
Nasiko's own product. ECC does not connect agents, start a server, enable
telemetry, share data, or operate that control plane. The bridge remains
experimental and opt-in.

Learn more in the Nasiko repository.

More reliable across every harness

• Native Antigravity 2.0: ECC installs to
  .agents/{rules,workflows,skills,agents} and safely migrates legacy ownership.
• Reliable uninstall: repeated selective installs keep a complete ownership
  ledger, so unchanged ECC-managed files can be removed cleanly.
• Safer migrations: OpenCode uses ~/.config/opencode, while OpenCode and
  Codex preserve edited or untracked files during cleanup.
• More workflows: Pi, Itô, agent evaluation, multi-model council, dev-team,
  living docs, secure terminal, TasteForge, and skill-comply all expand the
  shipped distribution.

ECC 2.2 ships 68 agents, 286 skills, and 94 command shims, plus the rules,
hooks, memory, continuous-learning, and installation surfaces around them.

Release assurance

The tagged commit must match origin/main. One npm archive is packed, hashed,
and tested on all three operating systems, then published under a staging tag.
Registry integrity must match before latest moves. The prior 2.1.0 package
remains the immediate rollback target.

Install or upgrade

For the native Claude Code plugin:

/plugin marketplace add https://github.com/affaan-m/ECC
/plugin install ecc@ecc

Existing Claude installs can run /plugin update ecc.

Thank you

A special thank you to returning contributors
@gaurav0107,
@Seekers2001, and
@t0kubetsu for substantial work on learning
workflows, collaborative skills, and shared testing.

Among our first-time contributors,
@renan-tiberio brought Pi support,
@romanclaudersoai1 repaired
/evolve, and
@djpjronline-netizen hardened hook
parsing.

Eighteen people made their first ECC contribution in 2.2. Thank you to everyone
who tested the installer, reported issues, and helped harden the release across
harnesses and operating systems.

New contributors

• @AlbertChiu777 in #2483
• @djpjronline-netizen in #2405
• @shanujans in #2538
• @HouMinXi in #2609
• @agent-jeong in #2670
• @karimfarahat in #2462
• @romanclaudersoai1 in #2664
• @Peopleoftech in #2668
• @alejandorosumah-mansa in #2700
• @imkp1 in #2637
• @shivaG5 in #2446
• @vitalii-novikov in #2520
• @etisamhaq in #2364
• @phumchai1515-prog in #2723
• @cagrisolakoglu in #2618
• @brianjkelly in #2703
• @renan-tiberio in #2759
• @ronimuliawan in #2793

Join us in the ECC Discord.

Full changelog: v2.1.0...v2.2.0

ECC 2.1.0: Plan Canvas, Kimi Harness, and Self-Hosted Compute

ECC 2.1 turns plan review into a visual loop and opens the harness to self-hosted models. Plan Canvas lets you review agent plans in the browser. Point at the part you mean instead of retyping it in chat. A new Kimi Code install target and a verified Itô GPU path make ECC + open-source models a first-class setup, backed by our public sponsors: Moonshot AI (Kimi), Itô, and Atlas Cloud.

Plan Canvas: review plans by pointing, not retyping

Download the MP4 demo

/plan ends with a confirm gate, and until now that review was a wall of markdown in the terminal. Now the agent opens the plan in a loopback-only browser canvas:

• Click elements or select text to attach numbered annotations
• Chat with the agent from a side rail while it works in the terminal
• Approve plan / Request changes buttons map directly onto /plan's CONFIRM gate
• Mermaid diagrams, tables, and task lists render natively; file edits live-reload the page
• Model- and harness-agnostic: a plain CLI + JSON protocol (ecc-plan-canvas), no Claude-only dependency

Kimi harness: Moonshot AI partnership

ECC now installs directly into Kimi Code (--target kimi). Kimi Code discovers the installed .kimi/AGENTS.md instructions and .kimi/skills/ workflows natively:

bash ./install.sh --target kimi --profile minimal
npx ecc doctor --target kimi
kimi

Self-host on Itô GPUs

Run ECC against any self-hosted open-source model. If you need GPU capacity, Itô is ECC's preferred compute sponsor. The integration shipped guarded end to end:

• ecc ito find: opt-in bridge to the canonical Itô CLI that submits a live, authenticated RFQ (it does not reserve capacity)
• Guarded live node qualification and read-only compute handoff
• Credential-bearing CLI shims are rejected outright

The sponsorship link itself is passive: it does not invoke an RFQ, reserve capacity, provision compute, or configure serving. Managed inference through Itô is not live yet. Any GPU provider works, and ECC stays provider-agnostic.

Partners

Moonshot AI (Kimi), Itô, and Atlas Cloud are now public sponsors of ECC. The README documents a recommended self-host path: Itô for GPU capacity, a Kimi checkpoint served behind a compatible endpoint, then Kimi Code + ECC on top. Each choice remains separate and swappable.

Also in 2.1

• Hermes and OpenClaw install targets: two more harnesses join Claude Code, Codex, OpenCode, Cursor, Gemini, Zed, Copilot, and Kimi
• Codex ECC navigation guide: find the right skill/command surface from inside Codex
• GateGuard path exemptions (GATEGUARD_EXEMPT_GLOBS) and configurable instinct injection (count + confidence threshold)
• PostToolUse hooks consolidated into sync/async dispatchers, with fewer processes per tool call
• Supply-chain hardening: the installer runtime passes strict vetting, and the pre-commit secret scan now catches Anthropic API keys (sk-ant-...)
• A long tail of community fixes across OpenCode, Windows, bun lockfiles, the dashboard, and project detection

The catalog now stands at 67 agents, 281 skills, and 94 command shims (2.0.0 shipped 64/261/84), plus hooks, rules, memory, continuous learning, and AgentShield.

Install or upgrade

/plugin marketplace add https://github.com/affaan-m/ECC
/plugin install ecc@ecc

Existing installs: /plugin update ecc

Community

Join the ECC community for release announcements, questions, and Show and Tell:
https://discord.gg/36yGMHGFbR

Full changelog: v2.0.0...v2.1.0

New Contributors

• @legeZZZ made their first contribution in #2200
• @tongshu2023 made their first contribution in #2216
• @mohameddsh3ban made their first contribution in #2195
• @fiedler-itlabs made their first contribution in #2240
• @daiki75 made their first contribution in #2245
• @wellkilo made their first contribution in #2194
• @rameshvr made their first contribution in #2144
• @aaronjmars made their first contribution in #2185
• @orbisai0security made their first contribution in #2149
• @xwang4-svg made their first contribution in #2186
• @kriptoburak made their first contribution in #2199
• @lamenting-hawthorn made their first contribution in #2235
• @mdayan8 made their first contribution in #2100
• @Naominour made their first contribution in #2188
• @Malphite10 made their first contribution in #2197
• @cogiwimute367-create made their first contribution in #2251
• @z3tz3r0 made their first contribution in #2250
• @rockwellwindsor made their first contribution in #2258
• @stroland02 made their first contribution in #2202
• @hretheum made their first contribution in #2221
• @BERORINPO made their first contribution in #2234
• @Victor-Casado made their first contribution in #2236
• @itkdm made their first contribution in #2241
• @leoeletronics made their first contribution in #2267
• @Seekers2001 made their first contribution in #2118
• @lucaszhu-hue made their first contribution in #2279
• @kapil971390 made their first contribution in #2292
• @danielnguyenfinhub made their first contribution in #2342
• @gpitrella made their first contribution in #2355
• @Angad2005 made their first contribution in #2366
• @quadcent made their first contribution in #2133
• @SiaoZeng made their first contribution in #2383
• @Tahiti18 made their first contribution in #2307
• @phobicdotno made their first contribution in #2343
• @sshworld made their first contribution in #2358
• @cd1amond made their first contribution in #2346
• @Tanaka-VivereGratis made their first contribution in #2388
• @KyawZinLatt made their first contribution in #2319
• @dieudonneAwa made their first contribution in #2288
• @carloscarvallo made their first contribution in #2273
• @mc856 made their first contribution in #2274
• @jvirgovic made their first contribution in #2338
• @m18897829375 made their first contribution in #2336
• @jack-finance-able made their first contribution in #2390
• @yerros made their first contribution in #2275
• @YuhaoLin2005 made their first contribution in #2377
• @Cb2i made their first contribution in #2437
• @28winz-bot made their first contribution in #2422
• @sutazca made their first contribution in #2425
• @nidelson made their first contribution in #2420
• @juujb made their first contribution in #2416
• @Zalasyu made their first contribution in #2432
• @Lxcardoza993 made their first contribution in #2433
• @haelyra made their first contribution in #2467
• @thejesh23 made their first contribution in #2515
• @nankingjing made their first contribution in #2501
• @samartomar made their first contribution in #2503
• @fletcherm-hub made their first contribution in #2460
• @someiyoshino-lab made their first contribution in #2471
• @WinterSold1er made their first contribution in #2511
• @EmadDoughan made their first contribution in #2543
• @Oleksandr-Kh4 made their first contribution in #2529
• @latreon made their first contribution in #2536

Full Changelog: v2.0.0...v2.1.0

ECC 2.0.0 is the stable graduation of the 2.0 line: ECC as a cross-harness operating system for agentic work. Claude Code stays first-class; Codex, OpenCode, Cursor, Gemini, Zed, and terminal-only workflows share the same skills, rules, hooks, MCP conventions, release gates, and operator workflows. This is the months-in-the-making release the rc.1 candidate was building toward.

The control pane (early build)

The ECC 2.0 operating surface, running locally (node scripts/control-pane.js): operator recall search, live session metrics, a work-items board with ready/running/blocked/done lanes, and an operator column that drives knowledge sync/recall, graph backfill, and the TUI. The session adapters and MCP inventory below feed this board. Build-in-public continues in the Discord #control-pane channel.

What is new

• 261 public skills across coding, research, security, media, enterprise ops, and agent workflows (rc.1 shipped 243) — plus 64 agents and 84 commands.
• The control-pane substrate — the foundation of the ECC 2.0 operating surface:
  • harness-neutral session adapters (ecc.session.v1) covering Claude Code, Codex, OpenCode, and dmux — one schema for "which agent is where, doing what"
  • MCP inventory (ecc.mcp.v1) — one normalized view of every MCP server config across harnesses, with fragmentation/drift detection and secret redaction (it caught a real arg-carried key leak during development)
  • worktree-lifecycle service — deterministic merge-conflict prediction and safe GC for parallel agent worktrees
• orch-* orchestrator skill family plus dynamic workflow team orchestration — multi-agent fan-out as a first-class surface.
• Single-connector MCP default policy — the default set is now exactly one connector (chrome-devtools); the other six retired to opt-in after a June 2026 audit. Policy + per-connector rationale: docs/MCP-CONNECTOR-POLICY.md.
• Rollout-derived optimization pack: parallel-execution-optimizer, benchmark-optimization-loop, data-throughput-accelerator, latency-critical-systems, recursive-decision-ledger.

Hardening since rc.1 (~30 PRs)

• Plugin hooks were silently no-ops on Node 21+ (#2184) — the hook runner depended on require.main under node -e; every plugin hook exited cleanly without running. If you are on Node 21+, update now.
• Windows reliability: CLAUDE_PLUGIN_ROOT normalization (#2139), stdin prompt passing (#2174), symlink/chmod test guards (#2171, #2176).
• Security: curl credentials out of argv (#2175), gateguard destructive-checkout gating (#2158) + env knobs (#2161), advisory intake hardening.
• Correctness: session-end $-sequence corruption (#2180), project-detect boundary matching (#2181), install manifest gaps (#2172), legacy shim truncation (#2167).
• Slimmer defaults: smaller OpenCode surface (#2140), rules/zh out of the default install (#2170).

The ECC Discord is live

https://discord.gg/36yGMHGFbR

• #announcements — releases auto-post and pin there (this very release was announced by the workflow shipped in it, #2201)
• #pr-and-issues — live feed of every PR, issue, and release from this repo
• the ECC bot (built in this release, dependency-free gateway client) answers in-server:
  • /skill name:<query> — look up any of the 261 skills
  • /docs query:<terms> — search the docs
  • /release — latest release
  • /ecc, /help
• #feedback / #feature-requests — read directly by the maintainer; #control-pane — build-in-public on the ECC 2.0 operating surface
• 16 channels, 6 categories, community roles for Contributors, ECC Tools users, and AgentShield users

Install or upgrade

/plugin marketplace add https://github.com/affaan-m/ECC
/plugin install ecc

Existing installs: /plugin update ecc

Download

• Release tarball: ecc-universal-2.0.0.tgz (attached below)
• Full changelog: v2.0.0-rc.1...v2.0.0

Verification

This release passed the full local validation gate before tagging: unicode safety, agent/command/rule/skill/hook validators, install-manifest checks, catalog and command-registry checks, and the complete test suite (SUITE EXIT: 0), plus markdownlint and the pre-push verification gate on every commit.

Credits

Authored by @affaan-m, built with Claude Code — including this release's Discord community, bot, and announce automation, which shipped themselves end to end.

ECC 2.0.0-rc.1

ECC 2.0.0-rc.1 is the first release-candidate surface for ECC as a cross-harness operating system for agentic work.

Claude Code remains a first-class target. Codex, OpenCode, Cursor, Gemini, Zed, and terminal-only workflows are now treated as execution surfaces that can share the same skills, rules, hooks, MCP conventions, release gates, and operator workflows.

What is new

• 243 public skills packaged across coding, research, security, media, enterprise ops, and agent workflows.
• A rollout-derived optimization pack for agents that need measured speedups instead of one-shot prompting.
• A public teaser Itô prediction-market skill pack for read-only research, basket comparison, oracle-style market intelligence, and risk review. Itô API access stays gated and separate from ECC Tools billing.
• AgentShield and supply-chain follow-up surfaces after the Mini Shai-Hulud/TanStack campaign work.
• Cross-harness package surfaces for Claude Code, Codex, OpenCode, Cursor, Gemini, Zed, and terminal workflows.
• Release-readiness gates for preview packs, public-action approvals, release URL ledgers, operator readiness, discussion state, Linear roadmap coverage, and supply-chain checks.

New optimization skills

• parallel-execution-optimizer: splits work into safe parallel lanes, separate worktrees, concurrent reads, and mergeable outputs.
• benchmark-optimization-loop: turns vague speed goals into measured variants, baselines, promotion gates, and rollback notes.
• data-throughput-accelerator: pushes large scans and corpus work into indexed, checkpointed, manifest-driven pipelines.
• latency-critical-systems: profiles p50, p95, queue time, cache hit rate, cold starts, and freshness before changing runtime paths.
• recursive-decision-ledger: converts repetitive rollout prompting into reusable decision ledgers with prior winners and explicit promotion criteria.

Why this matters

ECC is no longer just a Claude Code config pack. The release candidate packages the reusable layer underneath agent work: skills, hooks, rules, MCP conventions, command surfaces, test gates, and operator discipline that can move across harnesses.

The goal is simple: stop rewriting the same prompts, stop relying on fragile local habits, and turn working agent behavior into portable infrastructure.

Download

• Release tarball: https://github.com/affaan-m/ECC/releases/download/v2.0.0-rc.1/ecc-universal-2.0.0-rc.1.tgz
• Full changelog: v1.10.0...v2.0.0-rc.1

Additional registry and directory surfaces will follow the remaining publication gates. For this RC, the GitHub prerelease and tarball are the public packaged artifact.

Verification

This release candidate passed the local validation gate used for the final packaging work:

• npm test: 2568 passing
• npm run lint: passing
• node scripts/ci/validate-skills.js: 243 skills
• node scripts/ci/validate-install-manifests.js: 31 modules, 75 components, 6 profiles
• node scripts/ci/validate-no-personal-paths.js: passing
• npm run preview-pack:smoke -- --format json: passing
• npm pack --dry-run: includes the new optimization and Itô skill packs

RC boundary

This is a release candidate, not the final GA claim. Secrets, private operator state, OAuth tokens, personal datasets, and unsanitized local automations are not part of the public package.

ECC 1.10.0

What This Release Focuses On

• Harness reliability and cross-platform compatibility
• Eval-driven quality improvements
• Better workflow and operator ergonomics

Package Notes

• npm package: ecc-universal
• Claude marketplace/plugin identifier: everything-claude-code@everything-claude-code

New Contributors

• @tupe12334 made their first contribution in #649
• @ihimanss made their first contribution in #548
• @macstarosielec made their first contribution in #716
• @pvgomes made their first contribution in #736
• @berkcangumusisik made their first contribution in #744
• @nehaaprasad made their first contribution in #748
• @massimotodaro made their first contribution in #729
• @fischperson0000000 made their first contribution in #760
• @tae1344 made their first contribution in #956
• @Lidang-Jiang made their first contribution in #977
• @KT-lcz made their first contribution in #931
• @seancheick made their first contribution in #970
• @dagecko made their first contribution in #987
• @Infiniteyieldai made their first contribution in #986
• @senoldogann made their first contribution in #960
• @anuragg-saxenaa made their first contribution in #952
• @Ashwinhegde19 made their first contribution in #950
• @ToniDonDoni made their first contribution in #896
• @xingzihai made their first contribution in #929
• @haibindev made their first contribution in #911
• @techiro made their first contribution in #897
• @Yumerain made their first contribution in #855
• @sliver2er made their first contribution in #858
• @danielpetrica made their first contribution in #923
• @likzn made their first contribution in #985
• @694344851 made their first contribution in #818
• @sreedhargs89 made their first contribution in #959
• @drkeyurpatel-wq made their first contribution in #955
• @up2itnow0822 made their first contribution in #893
• @Xabilimon1 made their first contribution in #920
• @ohashi-mizuki made their first contribution in #1004
• @AndriyKalashnykov made their first contribution in #1007
• @singularityjason made their first contribution in #854
• @mike-batrakov made their first contribution in #1015
• @QWsin made their first contribution in #1019
• @eamanc-lab made their first contribution in #850
• @dani-mezei made their first contribution in #831
• @shreyas-lyzr made their first contribution in #833
• @2hxst made their first contribution in #1032
• @Millectable made their first contribution in #1042
• @Apptah made their first contribution in #1035
• @Divya-Somashekar made their first contribution in #1034
• @shahar-caura made their first contribution in #1052
• @kuqili made their first contribution in #1054
• @jayll1303 made their first contribution in #1025
• @bencmd88 made their first contribution in #1041
• @haochen806 made their first contribution in #1029
• @herakles-dev made their first contribution in #1036
• @Qingzhou-Joshua made their first contribution in #1038
• @cskwork made their first contribution in #741
• @yetval made their first contribution in #746
• @nayanjaiswal1 made their first contribution in #726
• @supmo668 made their first contribution in #848
• @hidai25 made their first contribution in #828
• @oxgeneral made their first contribution in #559

Full Changelog: v1.9.0...v1.10.0

Highlights

ECC Tools Pro is live. The GitHub App now has Stripe billing with a Pro tier ($19/seat/month) for private repo analysis, 50 analyses/month, and AgentShield-backed scanning. Free tier remains at 10 analyses/month for public repos. Install at github.com/apps/ecc-tools, upgrade at ecc.tools/pricing.

Selective install is here. Install only what you need with --with and --without flags. ecc install --profile developer --with lang:typescript --with agent:security-reviewer --without skill:continuous-learning. Agent and skill component families are now in the install manifest.

12 language ecosystems. C#, Rust, Java, Kotlin, C++, Go, Python, TypeScript, Perl, PyTorch, Nuxt 4, and Flutter now have dedicated rules, agents, or skills.

AgentShield v1.4.0 ships alongside. False positive scanning fixed. CVE database with 25+ known MCP vulnerabilities. Supply chain verification. Runtime monitoring. Watch mode. PR security gate. Org-wide policy enforcement.

What's New

Agents (28 total, +9 new)

• typescript-reviewer — TypeScript/JavaScript code review
• pytorch-build-resolver — PyTorch build error resolution
• java-build-resolver — Maven/Gradle build resolution
• java-reviewer — Java code review
• rust-reviewer — Rust code review
• flutter-reviewer — Flutter/Dart code review
• harness-optimizer — Cross-harness optimization
• kotlin-build-resolver — Kotlin build resolution
• kotlin-reviewer — Kotlin code review

Skills (116 total, +15 new)

• pytorch-patterns — PyTorch best practices
• nuxt4-patterns — Nuxt 4 frontend patterns
• codebase-onboarding — Onboarding new contributors
• architecture-decision-records — ADR workflow
• agent-eval — Head-to-head coding agent comparison
• documentation-lookup — Doc retrieval patterns
• bun-runtime — Bun runtime patterns
• nextjs-turbopack — Next.js Turbopack patterns
• Plus 7 more language-specific skills

Rules (6 new language packs)

• C# (coding-style, patterns, security, testing, hooks)
• Rust, Java, C++, Perl language rules
• Flutter/Dart rules

Infrastructure

• Selective install with --with/--without CLI flags
• Agent description compression with lazy loading
• Skill inspection logic for recurring failure pattern detection
• Governance event capture hooks
• MCP health-check hook with auto-reconnect
• SQLite state store for session/skill/decision tracking
• Session adapters for canonical snapshots
• Codex sync merges AGENTS.md instead of replacing

Bug Fixes

• Windows CI fully green (all 38 matrix jobs passing)
• SessionStart ANSI escape code sanitization
• Observer memory explosion with throttling
• Flaky observer bash test on Windows skipped
• Catalog count validation fixed
• Plugin root auto-detection fallback

ECC Tools App

• Stripe billing live (Pro $19/seat, Enterprise contact)
• GitHub OAuth checkout flow
• Analyzer quality improvements (7 greptile findings fixed)
• Duplicate PR prevention with idempotency
• Auto-analysis skip list for managed repos
• Push cooldown increased to 7 days

AgentShield (v1.4.0)

• False positive fix: deny rules and block hooks no longer flagged as threats
• CVE database with 25+ known MCP vulnerabilities
• MCP tool description poisoning detection (5 new rules)
• Continuous scanning with config drift detection (--watch mode)
• Runtime monitoring daemon for tool calls
• Supply chain verification for MCP npm packages
• PR security gate with baseline comparison
• Organization-wide security policy enforcement
• CLAUDE.md filesystem permission check
• 1,609 tests passing

Stats

• 219 commits since v1.8.0
• 30+ contributors
• 88K+ GitHub stars
• 1,540+ tests passing across all platforms

Install / Upgrade

npx claudepluginhub affaan-m/everything-claude-code --plugin everything-claude-code

Or install manually in Claude Code.

ECC v1.8.0: Harness Performance Release

v1.8.0 shifts ECC from "config bundle" to a practical agent harness performance system: more reliable hooks, stronger eval/quality controls, and tighter parity across Claude Code, Cursor, OpenCode, and Codex (macOS app + CLI).

What’s New

• New harness commands
  • /harness-audit
  • /loop-start
  • /loop-status
  • /quality-gate
  • /model-route
• NanoClaw v2 upgrades
  • Model routing
  • Skill hot-load
  • Session branch/search/export/compact/metrics
• New harness skills and operators
  • agent-harness-construction
  • agentic-engineering
  • ralphinho-rfc-pipeline
  • ai-first-engineering
  • enterprise-agent-ops
  • nanoclaw-repl
  • continuous-agent-loop
• New agents
  • harness-optimizer
  • loop-operator

Reliability and Hook Fixes

• Fixed SessionStart root resolution (robust fallback when plugin root env is missing).
• Fixed empty/partial session summary behavior by moving summary persistence to the Stop lifecycle path where transcript payload exists.
• Added explicit hook controls:
  • ECC_HOOK_PROFILE=minimal|standard|strict
  • ECC_DISABLED_HOOKS=comma,separated,ids
• Replaced fragile inline one-liner hook commands with dedicated scripts for safer cross-platform behavior.
• Improved Windows path handling for doc-warning/whitelist logic.
• Hardened observer behavior for non-interactive environments to reduce hang risk.

Cross-Harness Parity

• Claude Code, Cursor, OpenCode, and Codex now share tighter behavior around hooks, commands, and skill execution semantics.
• OpenCode command mirrors added for the new harness command set.
• Codex config/docs alignment improved for both app and CLI workflows.

Compatibility Notes

• autonomous-loops is retained for compatibility in this release.
• continuous-agent-loop is now the canonical skill name.

Upgrade Path

1. Pull latest and reinstall/update ECC.
2. Run /harness-audit to baseline reliability and risk.
3. Use /quality-gate . --strict on active repos.
4. If needed, tune hooks with ECC_HOOK_PROFILE and ECC_DISABLED_HOOKS.

Credits

• frontend / marketing slides skill inspired by @zarazhangrui
• continuous learning skill - homunculus-inspired by @humanplane
• slop guards inspired by @alexfazio

Supporting Docs

• Release notes source: docs/releases/1.8.0/release-notes.md
• Launch assets: docs/releases/1.8.0/

ECC v1.7.0 — Cross-Platform Expansion & Presentation Builder

992 tests. 56 skills. 33 commands. 14 agents. 4 major coding harnesses.

This release tightens ECC as a single source of truth across Claude Code, Codex, Cursor, and OpenCode, adds a full presentation-building workflow, and expands the repo with reusable content, research, and fundraising skills that are generic enough for any team to use.

Highlights

• Direct Codex support — ECC now ships cleaner Codex app + CLI support with AGENTS.md-based setup, installer support, and Codex-facing docs
• Cross-platform harness parity — Cursor, Codex, and OpenCode support were expanded so the same repo structure works cleanly across all major AI coding tools
• New frontend-slides skill — Build animation-rich, zero-dependency HTML slide decks and convert PPTX workflows into web-native presentations with strict viewport-fit rules
• 5 new generic business/content skills — article-writing, content-engine, market-research, investor-materials, and investor-outreach
• Cowork marketplace compatibility — Plugin manifests were reworked for Cowork compatibility and cleaner marketplace behavior
• Chief-of-staff agent — Added a communication-triage oriented agent for coordination-heavy workflows

Skills Added In This Cycle

Platform and Workflow Improvements

• Improved Codex packaging and discovery
• Better OpenCode package/release alignment
• Cross-platform hook and path fixes
• Auto-detect formatter support in the post-edit hook
• Cleaner rule priority and development workflow documentation
• Better skill metadata traceability and distribution support

Why This Matters

ECC is no longer just “a Claude Code plugin.” It now works as a multi-harness operating layer for modern AI coding environments while keeping the repo coherent enough to ship from one place.

Detailed Commit Log

• chore: release v1.7.0
• feat: add generic content and investor skills
• Add frontend slides skill across platforms
• fix: restructure plugin manifest for Cowork marketplace compatibility
• fix: sync plugin manifest to v1.6.0, add chief-of-staff agent, fix placeholder email
• fix(docs): use namespaced /everything-claude-code:plan in README examples
• fix(hooks): extract doc-warning hook to external script to fix CI
• feat(agents): add chief-of-staff communication triage agent (#280)
• feat: add Cursor, Codex, and OpenCode harnesses — maximize every AI coding tool
• feat: add security guides and sanitize external links across repo
• Merge pull request #233 from andydiaz122/nano_claw_v1
• Merge pull request #276 from pangerlkr/patch-7
• Merge pull request #273 from bintocher/bintocher/issue-246
• Merge pull request #255 from CindyPersonalGit/main
• Merge pull request #252 from pythonstrup/feat/auto-detect-formatter
• Merge pull request #250 from OkminLee/feat/skills/ios26-three-skills
• Merge pull request #244 from maxdimitrov/feat/rules/swift
• Merge pull request #242 from t-s-li/fix/observe-hook-phase-detection
• Merge pull request #277 from pangerlkr/docs/rule-priority-conflict-resolution
• Merge pull request #278 from pangerlkr/feat/separate-development-workflow-235
• fix: address CodeRabbit review — deduplicate prompt, fix skill count
• docs(hooks/README): update Doc file blocker to reflect warning-only behavior
• fix(hooks): address review feedback - fix brace escaping, cross-platform paths, update description
• feat(rules): add development-workflow.md with Feature Implementation Workflow (refs #235)
• refactor(rules): split Feature Implementation Workflow from git-workflow.md (refs #235)
• docs(rules): add Rule Priority section to resolve language-specific vs common rule conflicts
• Update hooks.jsonRefine Write hook: cross-platform path handling, skippability, and expanded whitelist. Addressing #264, #267, #275, and #248.
• Refine hooks for documentation file management
• fix(skills): improve code examples in iOS 26 skills
• fix(readme): correct skill count to 48 and add missing skill-stocktake to directory listing
• fix(skills): add Approachable Concurrency build settings note to PhotoProcessor example
• fix(skills): address code review feedback on iOS 26 skill examples
• chore: update skill count from 43 to 46, add 3 iOS 26 skills to directory listing
• feat(skills): add swift-concurrency-6-2 skill for Approachable Concurrency
• feat(skills): add foundation-models-on-device skill for Apple on-device LLM
• feat(skills): add liquid-glass-design skill for iOS 26 Liquid Glass UI system
• fix: add origin metadata to skills for traceability Add origin field to all skill files to track their source repository. This enables users to identify where distributed skills originated from. Fixes #246
• address review: remove .cursor/ duplicate, use is not None checks
• fix: use CLI argument for hook phase detection in observe.sh
• feat(skills): add visa-doc-translate skill
• feat: auto-detect formatter in post-edit hook (Biome/Prettier)
• fix: correct SwiftUI skill ViewModel injection and Equatable comparison
• feat: add swiftui-patterns skill
• feat: add Swift language-specific rules
• feat: add NanoClaw agent REPL — persistent session-aware CLI for ECC

ECC v1.6.0 — Codex Edition

309 commits. 30 community PRs merged. 978 tests. 4 platforms.

Everything Claude Code now supports Claude Code, Cursor, OpenCode, and Codex — every major AI coding tool from a single config source.

Highlights

• Codex Support — Full compatibility with OpenAI Codex CLI via new codex-setup command and instructions generator
• 7 New Skills — analyze-repo, skill-creator, instinct-creator, hook-creator, workflow-discovery, pattern-detection, learning-dna
• AgentShield Integration — 102 security rules, 912 tests across all protection layers
• GitHub Marketplace — Listed and installable at github.com/marketplace/ecc-tools
• Promo Video — Professional Remotion-rendered 1080p promo with real demo footage

New Commands

Bug Fixes

• Fixed agent count display (15→13 active agents)
• Fixed skill count (32→43 skills)
• Fixed command count (30→31 commands)
• Corrected contributor count across all docs
• Fixed markdown rendering in several skill outputs
• Resolved hook permission conflicts with Codex integration

Community

• 30 community PRs merged in this release cycle
• PRs #268, #265 merged; #272, #271 closed
• Active discussions on 10+ open issues
• Special thanks to all 44 contributors

Stats

"Your repository teaches itself to Claude Code."

Install: gh ext install affaan-m/everything-claude-code
Marketplace: github.com/marketplace/ecc-tools