ecl-harness-engineer
Create or audit ECL Agent Harness infrastructure: AGENTS.md, change tracking, repository guidance, lint checks, CI gates, and agent handoff docs.
Profile →npx skills add sickn33/agentic-awesome-skills --skill ecl-harness-engineer --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Designs and implements harness infrastructure to support AI agent collaboration within a repository. It targets creating or auditing repository-level artifacts such as AGENTS.md, ECL-related documentation, status handoff docs, change tracking mechanisms, and supporting scripts for linting and CI gates. The goal is to ensure a reliable environment where AI agents can operate with a stable, observable, and auditable harness rather than implementing business features directly.
How it works
Follows a unified workflow to detect gaps between the current state and a target harness state, then fills the gaps. It proceeds through: Quick Detection + Intent Confirmation, Analysis, Intake Review + Delta Synthesis, Creation/Update, and Verification + Handoff. It checks for the presence of harness components (AGENTS.md, docs/ECL.md, harness/changes, lint scripts, and CI scaffolding), evaluates the repository’s tech stack, and classifies project readiness (Empty, Code Only, Partial Harness, Harness Present). For ECL guidance, it handles small vs structured changes, creates or updates docs, templates, scripts, and CI examples, and enforces review and rollback discipline. Phase 2 emphasizes codebase understanding, adapter selection based on tech stack, and the appropriate command surface for scripts or CI tasks. It uses explicit commands and file names as defined in the workflow (e.g., AGENTS.md, docs/ECL.md, harness/changes, harness/templates, scripts/harness-change., scripts/harness-evolve.). It specifies that evaluation and long-term state features are optional profiles only when requested.
When to use it
Use when a repository needs AI-agent collaboration infrastructure (AGENTS.md, docs/ECL.md, STATUS doc, harness change tracking, or CI gates) or when auditing an existing harness for missing lifecycle docs, templates, lint checks, environment contracts, or CI integration. Do not use for ordinary feature work unless it concerns repository harness improvements.
What it can touch
The skill identifies and creates/updates repository artifacts and harness-related scripts. It recognizes and references artifacts such as AGENTS.md, docs/ECL.md, docs/STATUS.md, harness/changes, harness/templates/change, harness/templates/evolve, scripts/harness-change., scripts/harness-evolve., scripts/lint-*, Makefile, and language-specific adapters. It relies on existing project entrypoints (package.json, Makefile, or other scripts) to determine a suitable command surface and does not mandate a specific runtime beyond adapting to the detected tech stack. It can generate or modify documentation, lint rules,-change templates, and CI examples in alignment with the target project.
Caveats
The skill is limited to creating or auditing harness infrastructure and does not replace product requirements, feature implementation, or release approval. Generated docs, scripts, and CI examples must be adapted to the repository’s actual stack, security model, and contributor workflow before enforcement. Auto-evolve recommendations are guidance only and require review, validation, and rollback discipline. The skill emphasizes evidence-based changes rather than autonomous policy updates.
# ECL Harness Engineer Design and create Harness Engineering infrastructure so AI agents can work reliably in a codebase. > **Core Philosophy**: "Intelligence without infrastructure is just a demo." The Agent Harness is the Operating System — the LLM is just the CPU. The repository becomes the single source of truth — if an agent can't see it in context, it doesn't exist. ## When to Use This Skill - Use when a repository needs AI-agent collaboration infrastructure such as `AGENTS.md`, `docs/ECL.md`, `docs/STATUS.md`, harness change tracking, or mechanical validation gates. - Use when auditing an existing Agent Harness for missing ECL lifecycle docs, change templates, lint checks, environment contracts, or CI integration. - Use when converting repeated agent workflow failures into repository-local documentation, tests, lint rules, or lightweight auto-evolution checks. - Do not use for ordinary business feature implementation unless the requested work is specifically about creating or improving the repository harness. ## Limitations - This skill creates or audits harness infrastructure; it does not replace product requirements, implementation planning, code review, or release approva
- When to Use This Skill
- Limitations
- Unified Workflow
- Phase 1: Quick Detection + Intent Confirmation
- 1.1 Project State Detection
- 1.2 Classify Project State
- 1.3 Baseline Verification Snapshot
- 1.4 Intent Confirmation
- 1.5 ECL Work Intake Rules
- Phase 2: Analysis
- 2.1 Execution Mode
- 2.2 Project Identity Extraction
- 2.3 Adapter Selection
- 2.4 Command Surface Selection
Count files Check harness components Detect tech stack if test -f go.mod; then TECH="Go" elif test -f package.json; then TECH="TypeScript/Node.js" elif test -f requirements.txt || test -f pyproject.toml; then TECH="Python" else TECH="Unknown" fi cat harness/.analysis/architecture.json cat harness/.analysis/audit.json
What does the ecl-harness-engineer skill do?
Create or audit ECL Agent Harness infrastructure: AGENTS.md, change tracking, repository guidance, lint checks, CI gates, and agent handoff docs.
How do I install it?
Run `npx skills add sickn33/agentic-awesome-skills --skill ecl-harness-engineer --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From sickn33/agentic-awesome-skills, a repository with 44,414 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.