v1.8.0
ECC v1.8.0 — Harness Performance & Cross-Platform Reliability
Added
- Add /harness-audit command
- Add /loop-start command
- Add /loop-status command
- Add /quality-gate command
- Add /model-route command
- Add model routing to NanoClaw v2
- Add skill hot-load to NanoClaw v2
- Add session branch/search/export/compact/metrics to NanoClaw v2
- Add agent-harness-construction skill
- Add agentic-engineering skill
- Add ralphinho-rfc-pipeline skill
- Add ai-first-engineering skill
- Add enterprise-agent-ops skill
- Add nanoclaw-repl skill
- Add continuous-agent-loop skill
- Add harness-optimizer agent
- Add loop-operator agent
- Add ECC_HOOK_PROFILE environment variable with minimal, standard, or strict options
- Add ECC_DISABLED_HOOKS environment variable to disable comma-separated hook IDs
- Add explicit hook controls for safer configuration
Changed
- Tighten behavior parity across Claude Code, Cursor, OpenCode, and Codex for hooks, commands, and skill execution semantics
- Add OpenCode command mirrors for new harness command set
- Improve Codex config and docs alignment for app and CLI workflows
Fixed
- Fix SessionStart root resolution with robust fallback when plugin root env is missing
- Fix empty and partial session summary behavior by moving summary persistence to Stop lifecycle path
- Replace fragile inline one-liner hook commands with dedicated scripts for safer cross-platform behavior
- Improve Windows path handling for doc-warning and whitelist logic
- Harden observer behavior for non-interactive environments to reduce hang risk
Deprecated
- Retain autonomous-loops for compatibility but discontinue in favor of continuous-agent-loop as canonical skill name
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).
https://github.com/user-attachments/assets/18f05e41-a49b-4126-9771-3df1e07cd7b7
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-constructionagentic-engineeringralphinho-rfc-pipelineai-first-engineeringenterprise-agent-opsnanoclaw-replcontinuous-agent-loop
- New agents
harness-optimizerloop-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|strictECC_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-loopsis retained for compatibility in this release.continuous-agent-loopis now the canonical skill name.
Upgrade Path
- Pull latest and reinstall/update ECC.
- Run
/harness-auditto baseline reliability and risk. - Use
/quality-gate . --stricton active repos. - If needed, tune hooks with
ECC_HOOK_PROFILEandECC_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/