harness-skills
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Check it deeper on your machine
It has changed since its scan on 13 Nov 2026.
As a Claude Code plugin, install it and then ask Claude Code:
/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/harness/harness-skills with scanskillsafety
Or from a terminal, clone it and run it:
git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/harness/harness-skills'
It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.
Badge for your README
Markdown
[](/listings/harness--harness-skills)HTML
<a href="/listings/harness--harness-skills"><img src="/api/listings/harness--harness-skills/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Each category is its own plugin. On Claude Code 2.1.275 or later, install one in a single step (swap `harness-fme` for any plugin in the tables above; each table shows its own install line):
/plugin install harness-fme --marketplace harness/harness-skills
On older versions, add the marketplace first, then install:
/plugin marketplace add harness/harness-skills /plugin install harness-fme@harness-skills
The same from a shell:
claude plugin install harness-fme --marketplace harness/harness-skills # older versions: add the marketplace first claude plugin marketplace add harness/harness-skills claude plugin install harness-fme@harness-skills
Invoke an installed skill as `/<plugin>:<skill>`:
/harness-fme:explain-flag Explain the new_checkout flag in production
To update, run `/plugin marketplace update harness-skills`. Plugins are versioned by git commit. Each installed plugin caches a full copy of this repository (about 8 MB), because skills link to the shared `references/` and `templates/` at the repo root. Installing all nine plugins stores nine copies.
**Known gap:** plugin installs do not load the repo-level `AGENTS.md` rules (scope establishment, dependency checks, schema validation). FME skills load their shared rules through `references/fme/`; other skills rely on their own instructions. To get the repo-level rules, work from a clone (see [Quick Start without a plugin](#quick-start-without-a-plugin)).
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
.claude-plugin/marketplace.json · Manifest · 238 lines
- {
- "name": "harness-skills",
- "owner": {
- "name": "Harness"
- },
- "metadata": {
- "description": "74 Harness skills for AI coding agents, as 10 domain plugins."
- },
- "plugins": [
- {
- "name": "harness-pipelines",
- "displayName": "Harness Pipelines & Execution",
- "description": "9 skills: Create v0 and v1 pipelines, triggers, templates and freeze windows, then run, debug and migrate them.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/pipelines"
- ],
- "category": "ci-cd",
- "keywords": [
- "harness",
- "pipelines",
- "ci-cd",
- "triggers",
- "templates"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-infrastructure",
- "displayName": "Harness Infrastructure & Resources",
- "description": "5 skills: Define services, environments, infrastructure, connectors and secrets for Harness deployments.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/infrastructure"
- ],
- "category": "infrastructure",
- "keywords": [
- "harness",
- "services",
- "environments",
- "connectors",
- "secrets"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-access-control",
- "displayName": "Harness Access Control & Users",
- "description": "2 skills: Manage users, user groups, service accounts, roles, resource groups and RBAC assignments.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/access-control"
- ],
- "category": "security",
- "keywords": [
- "harness",
- "rbac",
- "users",
- "roles"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-fme",
- "displayName": "Harness FME (Feature Flags & Experimentation)",
- "description": "16 skills: Discover, create, target, roll out, clean up and experiment with Harness FME feature flags, segments and metrics.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/fme"
- ],
- "category": "feature-flags",
- "keywords": [
- "harness",
- "fme",
- "feature-flags",
- "experimentation",
- "split"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-database",
- "displayName": "Harness Database DevOps",
- "description": "1 skill: Generate, review and run Liquibase changesets with Harness Database DevOps.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/database"
- ],
- "category": "database",
- "keywords": [
- "harness",
- "database",
- "liquibase",
- "dbops"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-platform",
- "displayName": "Harness Platform Operations",
- "description": "8 skills: Operate delegates, IaCM workspaces, cloud development environments, artifact registries, SBOMs, supply chain and pull requests.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/platform"
- ],
- "category": "platform",
- "keywords": [
- "harness",
- "delegates",
- "iacm",
- "cde",
- "artifacts",
- "sbom"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-security",
- "displayName": "Harness Security Testing & Supply Chain",
- "description": "11 skills: Add STO code, secret, container and DAST scans, SLSA provenance and artifact signing, and manage exemptions and PR attestation.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/security"
- ],
- "category": "security",
- "keywords": [
- "harness",
- "sto",
- "security",
- "slsa",
- "signing"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-observability-governance",
- "displayName": "Harness Observability & Governance",
- "description": "19 skills: Costs, DORA, SLOs, chaos, scorecards, audits, OPA policies, IDP, SEI analytics and deployment risk insights.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/observability-governance"
- ],
- "category": "observability",
- "keywords": [
- "harness",
- "observability",
- "governance",
- "chaos",
- "slo",
- "opa"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-unified-data-platform",
- "displayName": "Harness Unified Data Platform",
- "description": "1 skill: Query Harness data with HQL: discover queryable types and fields, then write, validate and run queries.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/unified-data-platform"
- ],
- "category": "data",
- "keywords": [
- "harness",
- "hql",
- "analytics",
- "udp"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- },
- {
- "name": "harness-agents",
- "displayName": "Harness AI Agents",
- "description": "2 skills: Create Harness AI agents and agent templates for agentic pipeline workflows.",
- "source": ".",
- "strict": false,
- "skills": [
- "./skills/agents"
- ],
- "category": "ai-agents",
- "keywords": [
- "harness",
- "agents",
- "ai"
- ],
- "author": {
- "name": "Harness"
- },
- "repository": "https://github.com/harness/harness-skills",
- "license": "Apache-2.0"
- }
- ]
- }
README.md · README · 602 lines
- # Harness Skills
- Model-agnostic AI coding skills for the [Harness.io](https://harness.io) CI/CD platform. Generate pipeline YAML, manage resources, debug failures, analyze costs, and more from natural language.
- This repository is designed as a workflow system, not just a folder of prompts. The top-level instructions (`AGENTS.md`, with `CLAUDE.md` importing it for Claude Code, plus `.cursor/rules/harness.mdc` and `.github/copilot-instructions.md`) establish shared behavior when you work from a clone, while individual skills specialize in creation, debugging, governance, and reporting tasks. Plugin installs ship those instruction files but do not load them; see [Install as a Claude Code Plugin](#install-as-a-claude-code-plugin).
- ## Prerequisites
- - [Harness MCP v2 Server](https://github.com/harness/mcp-server) - required for MCP-powered skills. Most skills in this repo depend on it for Harness API access. It is not bundled with any plugin; see [MCP Setup](#mcp-setup).
- ## Available Skills
- The tables below are generated from `skills/categories.json` and each skill's frontmatter. Claude Code invokes a skill as `/<plugin>:<skill>`; Cursor, Codex, and other tools use the skill name.
- <!-- BEGIN GENERATED:skills -->
- <!-- Generated by scripts/generate_catalog.py from skills/categories.json and SKILL.md frontmatter. Do not edit. -->
- ### Harness Pipelines & Execution
- Plugin `harness-pipelines` (9 skills), folder `skills/pipelines/`.
- Install in Claude Code: `/plugin install harness-pipelines --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [create-pipeline](skills/pipelines/create-pipeline/SKILL.md) | `/harness-pipelines:create-pipeline` | Generate v0 pipeline YAML (CI, CD, combined, approvals) |
- | [create-pipeline-v1](skills/pipelines/create-pipeline-v1/SKILL.md) | `/harness-pipelines:create-pipeline-v1` | Generate v1 simplified pipeline YAML |
- | [create-template](skills/pipelines/create-template/SKILL.md) | `/harness-pipelines:create-template` | Create reusable step, stage, pipeline, and step group templates |
- | [create-trigger](skills/pipelines/create-trigger/SKILL.md) | `/harness-pipelines:create-trigger` | Create webhook, scheduled, and artifact triggers |
- | [debug-pipeline](skills/pipelines/debug-pipeline/SKILL.md) | `/harness-pipelines:debug-pipeline` | Diagnose pipeline execution failures |
- | [enable-code-coverage](skills/pipelines/enable-code-coverage/SKILL.md) | `/harness-pipelines:enable-code-coverage` | Onboard repositories to Harness Code Coverage with PR gating and baseline triggers |
- | [manage-freeze-windows](skills/pipelines/manage-freeze-windows/SKILL.md) | `/harness-pipelines:manage-freeze-windows` | Create, toggle, and manage deployment freeze windows and the project-level global freeze |
- | [migrate-pipeline](skills/pipelines/migrate-pipeline/SKILL.md) | `/harness-pipelines:migrate-pipeline` | Convert v0 pipelines to v1 format |
- | [run-pipeline](skills/pipelines/run-pipeline/SKILL.md) | `/harness-pipelines:run-pipeline` | Execute and monitor pipeline runs |
- ### Harness Infrastructure & Resources
- Plugin `harness-infrastructure` (5 skills), folder `skills/infrastructure/`.
- Install in Claude Code: `/plugin install harness-infrastructure --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [create-connector](skills/infrastructure/create-connector/SKILL.md) | `/harness-infrastructure:create-connector` | Create connectors (GitHub, AWS, GCP, Azure, Docker, K8s) |
- | [create-environment](skills/infrastructure/create-environment/SKILL.md) | `/harness-infrastructure:create-environment` | Create environments (PreProduction, Production) with overrides |
- | [create-infrastructure](skills/infrastructure/create-infrastructure/SKILL.md) | `/harness-infrastructure:create-infrastructure` | Define infrastructure (K8s, ECS, Serverless) |
- | [create-secret](skills/infrastructure/create-secret/SKILL.md) | `/harness-infrastructure:create-secret` | Manage secrets (SecretText, SecretFile, SSHKey, WinRM) |
- | [create-service](skills/infrastructure/create-service/SKILL.md) | `/harness-infrastructure:create-service` | Define services (Kubernetes, Helm, ECS) with artifact sources |
- ### Harness Access Control & Users
- Plugin `harness-access-control` (2 skills), folder `skills/access-control/`.
- Install in Claude Code: `/plugin install harness-access-control --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [manage-roles](skills/access-control/manage-roles/SKILL.md) | `/harness-access-control:manage-roles` | RBAC roles, assignments, permissions, and resource groups |
- | [manage-users](skills/access-control/manage-users/SKILL.md) | `/harness-access-control:manage-users` | Manage users, user groups, and service accounts |
- ### Harness FME (Feature Flags & Experimentation)
- Plugin `harness-fme` (16 skills), folder `skills/fme/`.
- Install in Claude Code: `/plugin install harness-fme --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [choose-metric](skills/fme/choose-metric/SKILL.md) | `/harness-fme:choose-metric` | Recommend primary/guardrail metrics for an experiment or rollout, judged against a good-metric checklist |
- | [cleanup-feature-flags](skills/fme/cleanup-feature-flags/SKILL.md) | `/harness-fme:cleanup-feature-flags` | Remove a launched FME flag from code and open a PR; archive is handed to manage-flag-lifecycle |
- | [create-experiment](skills/fme/create-experiment/SKILL.md) | `/harness-fme:create-experiment` | Creation-only entry point to the manage-experiments workflow |
- | [create-feature-flag](skills/fme/create-feature-flag/SKILL.md) | `/harness-fme:create-feature-flag` | Create an FME flag that follows project conventions, with safe-default definitions |
- | [create-metric](skills/fme/create-metric/SKILL.md) | `/harness-fme:create-metric` | Design and create an FME metric, with intent-to-config guidance, traffic type, events, cap, and owners |
- | [discover-feature-flags](skills/fme/discover-feature-flags/SKILL.md) | `/harness-fme:discover-feature-flags` | Inventory, targeting-pattern search, promotion-aware rollout report, and stale-flag audit (read-only) |
- | [explain-flag](skills/fme/explain-flag/SKILL.md) | `/harness-fme:explain-flag` | Explain one flag's purpose, treatments, and per-environment targeting, flagging inconsistencies (read-only) |
- | [feature-flag-onboarding](skills/fme/feature-flag-onboarding/SKILL.md) | `/harness-fme:feature-flag-onboarding` | Set up and verify a first feature flag in an existing project; resume partial onboarding |
- | [fme-pipeline](skills/fme/fme-pipeline/SKILL.md) | `/harness-fme:fme-pipeline` | Generate FME pipelines for rollout scenarios, gates, and flag lifecycle |
- | [instrument-feature-flag](skills/fme/instrument-feature-flag/SKILL.md) | `/harness-fme:instrument-feature-flag` | Gate application code through an extensible flag abstraction, after stack detection and clarification |
- | [instrument-metric](skills/fme/instrument-metric/SKILL.md) | `/harness-fme:instrument-metric` | Wire up a track() call for a metric's event, placed at the real outcome, and verify it arrives |
- | [manage-experiments](skills/fme/manage-experiments/SKILL.md) | `/harness-fme:manage-experiments` | Feature-flag experiment design/lifecycle; delegates metrics and states CLI update limitations |
- | [manage-flag-lifecycle](skills/fme/manage-flag-lifecycle/SKILL.md) | `/harness-fme:manage-flag-lifecycle` | Flag metadata, archive readiness, archive/unarchive, and delete |
- | [manage-segments](skills/fme/manage-segments/SKILL.md) | `/harness-fme:manage-segments` | Segment metadata, STANDARD membership, and RULE_BASED editor workflows with mandatory pre-write reference checks |
- | [review-experiment-results](skills/fme/review-experiment-results/SKILL.md) | `/harness-fme:review-experiment-results` | Explain experiment results, including winner, significance, guardrail impact, and data-quality caveats |
- | [update-flag-targeting](skills/fme/update-flag-targeting/SKILL.md) | `/harness-fme:update-flag-targeting` | Ramp, rules, individual targets, allocation, treatments, kill/restore, env-to-env copy, and initializing definitions |
- ### Harness Database DevOps
- Plugin `harness-database` (1 skill), folder `skills/database/`.
- Install in Claude Code: `/plugin install harness-database --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [dbops-changeset](skills/database/dbops-changeset/SKILL.md) | `/harness-database:dbops-changeset` | Generate, refine, review, and execute Liquibase changesets with Harness DBOPS |
- ### Harness Platform Operations
- Plugin `harness-platform` (8 skills), folder `skills/platform/`.
- Install in Claude Code: `/plugin install harness-platform --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [create-sbom](skills/platform/create-sbom/SKILL.md) | `/harness-platform:create-sbom` | Add SBOM Orchestration (generation/attestation) to existing pipelines |
- | [enforce-sbom](skills/platform/enforce-sbom/SKILL.md) | `/harness-platform:enforce-sbom` | Add SBOM Policy Enforcement (verify attestation + OPA policy sets) |
- | [manage-artifacts](skills/platform/manage-artifacts/SKILL.md) | `/harness-platform:manage-artifacts` | Artifact registry setup, security scanning, and cross-region replication |
- | [manage-cde](skills/platform/manage-cde/SKILL.md) | `/harness-platform:manage-cde` | Cloud development environments, workspace templates, and auto-hibernation |
- | [manage-delegates](skills/platform/manage-delegates/SKILL.md) | `/harness-platform:manage-delegates` | Monitor delegate health and manage registration tokens |
- | [manage-iacm](skills/platform/manage-iacm/SKILL.md) | `/harness-platform:manage-iacm` | Terraform workspaces, drift detection, cost estimation, and approval workflows |
- | [manage-pull-requests](skills/platform/manage-pull-requests/SKILL.md) | `/harness-platform:manage-pull-requests` | List, create, update, merge, review, and comment on Harness Code pull requests |
- | [manage-supply-chain](skills/platform/manage-supply-chain/SKILL.md) | `/harness-platform:manage-supply-chain` | SBOM generation, artifact signing, and supply chain policy enforcement |
- ### Harness Security Testing & Supply Chain
- Plugin `harness-security` (11 skills), folder `skills/security/`.
- Install in Claude Code: `/plugin install harness-security --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [approve-exempt](skills/security/approve-exempt/SKILL.md) | `/harness-security:approve-exempt` | Approve pending STO exemptions at current scope or elevate to Project, Org, or Account |
- | [configure-agent-pr-attestation](skills/security/configure-agent-pr-attestation/SKILL.md) | `/harness-security:configure-agent-pr-attestation` | Install and configure the Claude Code PR attestation hook for signed in-toto attestations to the Harness Evidence Vault |
- | [configure-container-scan](skills/security/configure-container-scan/SKILL.md) | `/harness-security:configure-container-scan` | Add container image scanning to pipelines using Harness STO scanners (Trivy, Grype, Snyk, and more) |
- | [configure-dast-scan](skills/security/configure-dast-scan/SKILL.md) | `/harness-security:configure-dast-scan` | Add DAST scanning to pipelines for runtime application security testing (Traceable, Burp Suite, ZAP, Nikto, Nmap) |
- | [configure-repo-scan](skills/security/configure-repo-scan/SKILL.md) | `/harness-security:configure-repo-scan` | Configure code scanning in pipelines with STO security scanners |
- | [configure-secret-scan](skills/security/configure-secret-scan/SKILL.md) | `/harness-security:configure-secret-scan` | Add secret detection scanning to pipelines to find exposed credentials and API keys (Gitleaks, Semgrep, Snyk, and more) |
- | [enforce-slsa](skills/security/enforce-slsa/SKILL.md) | `/harness-security:enforce-slsa` | Add SLSA Verification (verify attestation + OPA policy sets on provenance) to pipelines |
- | [exempt-vuln](skills/security/exempt-vuln/SKILL.md) | `/harness-security:exempt-vuln` | Create STO vulnerability exemptions (waivers) scoped to Project, Pipeline, or Target |
- | [generate-slsa](skills/security/generate-slsa/SKILL.md) | `/harness-security:generate-slsa` | Add SLSA Generation (provenance + optional Cosign attestation) to existing pipelines |
- | [sign-artifact](skills/security/sign-artifact/SKILL.md) | `/harness-security:sign-artifact` | Add Artifact Signing (Cosign sign + optional .sig upload) to existing pipelines (SSCA) |
- | [verify-sign](skills/security/verify-sign/SKILL.md) | `/harness-security:verify-sign` | Add Artifact Verification (verify Cosign signature) to existing pipelines (SSCA) |
- ### Harness Observability & Governance
- Plugin `harness-observability-governance` (19 skills), folder `skills/observability-governance/`.
- Install in Claude Code: `/plugin install harness-observability-governance --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [ai-operations](skills/observability-governance/ai-operations/SKILL.md) | `/harness-observability-governance:ai-operations` | Predictive failure analysis and intelligent alert correlation (AIDA) |
- | [analyze-costs](skills/observability-governance/analyze-costs/SKILL.md) | `/harness-observability-governance:analyze-costs` | Cloud cost analysis, recommendations, and anomaly detection |
- | [audit-report](skills/observability-governance/audit-report/SKILL.md) | `/harness-observability-governance:audit-report` | Audit trails and compliance evidence (SOC2, GDPR, HIPAA) |
- | [chaos-code-risk-scan](skills/observability-governance/chaos-code-risk-scan/SKILL.md) | `/harness-observability-governance:chaos-code-risk-scan` | Scan a code repository for chaos-provable resilience risks and report them as a risk scan |
- | [chaos-dr-test](skills/observability-governance/chaos-dr-test/SKILL.md) | `/harness-observability-governance:chaos-dr-test` | Create and edit DR Test pipelines with chaos steps |
- | [chaos-experiment](skills/observability-governance/chaos-experiment/SKILL.md) | `/harness-observability-governance:chaos-experiment` | Create, edit, and reference chaos experiments (faults, probes, actions) |
- | [create-policy](skills/observability-governance/create-policy/SKILL.md) | `/harness-observability-governance:create-policy` | Create OPA governance policies for supply chain security |
- | [deployment-readiness](skills/observability-governance/deployment-readiness/SKILL.md) | `/harness-observability-governance:deployment-readiness` | Pre-deployment readiness checks, environment drift, and canary decisions |
- | [dora-metrics](skills/observability-governance/dora-metrics/SKILL.md) | `/harness-observability-governance:dora-metrics` | DORA metrics and engineering performance reports |
- | [gitops-status](skills/observability-governance/gitops-status/SKILL.md) | `/harness-observability-governance:gitops-status` | GitOps application health, sync status, and pod logs |
- | [incident-response](skills/observability-governance/incident-response/SKILL.md) | `/harness-observability-governance:incident-response` | Deployment-incident correlation, blast radius analysis, and postmortems |
- | [manage-idp](skills/observability-governance/manage-idp/SKILL.md) | `/harness-observability-governance:manage-idp` | Service catalog, self-service workflows, documentation, and developer onboarding (IDP) |
- | [manage-slos](skills/observability-governance/manage-slos/SKILL.md) | `/harness-observability-governance:manage-slos` | SLO definition, error budgets, incident detection, and operational runbooks |
- | [optimize-pipeline](skills/observability-governance/optimize-pipeline/SKILL.md) | `/harness-observability-governance:optimize-pipeline` | Pipeline speed optimization, parallel testing, caching, and monorepo builds |
- | [pr-analysis](skills/observability-governance/pr-analysis/SKILL.md) | `/harness-observability-governance:pr-analysis` | PR pipeline impact, security review, and PR-to-production tracking |
- | [scorecard-review](skills/observability-governance/scorecard-review/SKILL.md) | `/harness-observability-governance:scorecard-review` | IDP scorecards and service maturity review |
- | [security-report](skills/observability-governance/security-report/SKILL.md) | `/harness-observability-governance:security-report` | Security vulnerabilities, SBOMs, and compliance reports |
- | [sei-analytics](skills/observability-governance/sei-analytics/SKILL.md) | `/harness-observability-governance:sei-analytics` | Sprint analytics, investment allocation, capacity forecasting, and release readiness |
- | [template-usage](skills/observability-governance/template-usage/SKILL.md) | `/harness-observability-governance:template-usage` | Template dependency tracking, impact analysis, and adoption |
- ### Harness Unified Data Platform
- Plugin `harness-unified-data-platform` (1 skill), folder `skills/unified-data-platform/`.
- Install in Claude Code: `/plugin install harness-unified-data-platform --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [write-hql](skills/unified-data-platform/write-hql/SKILL.md) | `/harness-unified-data-platform:write-hql` | Discover fields, then write, validate, and run dashboard-ready HQL queries |
- ### Harness AI Agents
- Plugin `harness-agents` (2 skills), folder `skills/agents/`.
- Install in Claude Code: `/plugin install harness-agents --marketplace harness/harness-skills`.
- | Skill | Invoke in Claude Code | Description |
- |-------|------------------------|-------------|
- | [create-agent](skills/agents/create-agent/SKILL.md) | `/harness-agents:create-agent` | Create and update Harness AI agent instances for agentic pipeline workflows |
- | [create-agent-template](skills/agents/create-agent-template/SKILL.md) | `/harness-agents:create-agent-template` | Generate AI agent templates (metadata.json, pipeline.yaml, wiki.MD) |
- <!-- END GENERATED:skills -->
- ## Install as a Claude Code Plugin
- Each category is its own plugin. On Claude Code 2.1.275 or later, install one in a single step (swap `harness-fme` for any plugin in the tables above; each table shows its own install line):
- ```
- /plugin install harness-fme --marketplace harness/harness-skills
- ```
- On older versions, add the marketplace first, then install:
- ```
- /plugin marketplace add harness/harness-skills
- /plugin install harness-fme@harness-skills
- ```
- The same from a shell:
- ```bash
- claude plugin install harness-fme --marketplace harness/harness-skills
- # older versions: add the marketplace first
- claude plugin marketplace add harness/harness-skills
- claude plugin install harness-fme@harness-skills
- ```
- Invoke an installed skill as `/<plugin>:<skill>`:
- ```
- /harness-fme:explain-flag
- Explain the new_checkout flag in production
- ```
- To update, run `/plugin marketplace update harness-skills`. Plugins are versioned by git commit. Each installed plugin caches a full copy of this repository (about 8 MB), because skills link to the shared `references/` and `templates/` at the repo root. Installing all nine plugins stores nine copies.
- **Known gap:** plugin installs do not load the repo-level `AGENTS.md` rules (scope establishment, dependency checks, schema validation). FME skills load their shared rules through `references/fme/`; other skills rely on their own instructions. To get the repo-level rules, work from a clone (see [Quick Start without a plugin](#quick-start-without-a-plugin)).
- ## Install as a Cursor Plugin
- Cursor installs the whole repository as one plugin, `harness-skills`, containing every skill in the tables above. Invoke a skill as `/<skill>`, for example `/explain-flag`.
- - **Team and Enterprise:** in the Cursor dashboard, open Plugins & MCPs, choose Add Marketplace, and import the GitHub repository `harness/harness-skills`.
- - **Local install:** clone into Cursor's local plugin folder:
- ```bash
- git clone https://github.com/harness/harness-skills.git ~/.cursor/plugins/local/harness-skills
- ```
- - **Public marketplace:** a listing in the public Cursor marketplace requires a submission at cursor.com/marketplace/publish. That has not been done yet.
- Cursor reads `.cursor-plugin/marketplace.json` before `.claude-plugin/marketplace.json`, so it sees the single `harness-skills` plugin rather than the nine Claude Code plugins.
- ## Install as a Codex Plugin
- Codex also installs the whole repository as one plugin, `harness-skills`:
- ```bash
- codex plugin marketplace add harness/harness-skills
- codex
- ```
- In Codex, open `/plugins`, install `harness-skills`, and start a new session.
- To enable it for one repository only, add this to `.codex/config.toml` in that repository:
- ```toml
- [plugins."harness-skills@harness-skills"]
- enabled = true
- ```
- Codex refreshes an installed plugin only when its version changes. The version is `version` in `skills/categories.json`, so Codex users do not see new skill content until that version is bumped.
- ## Quick Start without a plugin
- Clone the repository to use the skills, the shared playbooks, and the repo-level rules together in any coding agent:
- ```bash
- git clone https://github.com/harness/harness-skills.git
- ```
- Skill files live at `skills/<category>/<skill>/SKILL.md`, for example `skills/pipelines/create-pipeline/SKILL.md`.
- ### FME with CLI or MCP (any coding agent)
- 1. Keep the repository's `skills/`, shared `references/`, and `templates/` available together. Load `AGENTS.md` and the selected skill; referencing just its SKILL.md without the linked playbooks is insufficient.
- 2. Choose the transport available to your agent: a preconfigured Harness CLI with shell access, or the [Harness MCP server](https://github.com/harness/mcp-server). Confirm the account, organization, project and intended environments without exposing credentials. CLI-only FME sessions do not require MCP; authentication should use the tool's supported secure setup outside the prompt.
- 3. Read the [FME tool map](references/fme/tool-map.md#transport-preflight-and-handoffs) and check capabilities for the installed version. It documents unsupported operations; CLI/MCP support is not identical. Segment key operations require MCP; some experiment updates also require MCP.
- 4. Use the skill file as instructions even if the host has no slash-skill invocation. Load linked skills explicitly during handoffs. Code-editing workflows also need the application repository and its test tools; live verification additionally needs an authorized running application.
- The MCP examples in [MCP Setup](#mcp-setup) apply to MCP-powered sessions, not as a prerequisite for every FME workflow. See the [MCP setup playbook](references/mcp-setup.md) for connection, authentication, and read-only verification guidance.
- ### Claude Code
- Run Claude Code from the clone; project rules load via `CLAUDE.md` -> `@AGENTS.md`:
- ```bash
- cd harness-skills
- claude
- ```
- To invoke the skills as slash commands, add the clone as a local marketplace and install the plugins you need:
- ```
- /plugin marketplace add ./
- /plugin install harness-pipelines@harness-skills
- ```
- Then invoke skills by plugin and name:
- ```
- /harness-pipelines:create-pipeline
- Create a CI pipeline for a Node.js app that builds, runs tests,
- and pushes a Docker image to ECR
- ```
- ### Cursor
- 1. Clone the repo into your project or as a reference workspace.
- 2. The repo includes `.cursor/rules/harness.mdc`, which carries the same project rules as `AGENTS.md`.
- 3. Configure the Harness MCP server as described in [MCP Setup](#mcp-setup).
- 4. Reference individual skills in your prompts using `@file`:
- ```
- @harness-skills/skills/pipelines/create-pipeline/SKILL.md
- Create a CI pipeline for my Go service
- ```
- ### OpenAI Codex
- 1. Clone the repo into your working directory.
- 2. The repo includes `AGENTS.md` at the root, which Codex reads as project instructions.
- 3. Configure the Harness MCP server as described in [MCP Setup](#mcp-setup).
- 4. Reference individual skill files as context when prompting:
- ```
- Using the instructions in harness-skills/skills/pipelines/debug-pipeline/SKILL.md,
- diagnose why my deploy pipeline failed
- ```
- ### GitHub Copilot
- 1. Clone the repo into your project.
- 2. The repo includes `.github/copilot-instructions.md`, which Copilot automatically reads as project-level context in both GitHub.com and VS Code.
- 3. For VS Code, configure the Harness MCP server as described in [MCP Setup](#mcp-setup).
- 4. Reference skill files in Copilot Chat using `#file`:
- ```
- #file:harness-skills/skills/pipelines/create-pipeline/SKILL.md
- Create a CI pipeline for my Python service
- ```
- 5. For GitHub Copilot on GitHub.com, attach skill files as context in Copilot Chat or add them as knowledge base references in your Copilot organization settings.
- ### Windsurf / Other AI Editors
- The skills in this repo are plain Markdown files with YAML frontmatter. They work with any AI coding tool that supports:
- 1. **System instructions** - Use `AGENTS.md` as project-level context (`CLAUDE.md` imports it for Claude Code).
- 2. **Resource access** - Connect the [Harness MCP server](https://github.com/harness/mcp-server), or use the Harness CLI for supported FME operations.
- 3. **File context** - Reference individual `SKILL.md` files under `skills/<category>/<skill>/` in prompts.
- ### npx skills add
- `npx skills add harness/harness-skills` copies only the skill folders, so skills that link to the shared `references/` and `templates/` at the repo root lose those files. Prefer a plugin install or a clone until bundling ships.
- ## MCP Setup
- The Harness MCP server is not bundled with any plugin; configure it for each tool you use. The examples run the server locally with `npx`. Never commit a real API key: use an environment reference or a placeholder like `<your-api-key>`. See the [MCP setup playbook](references/mcp-setup.md) for hosted vs. local servers, authentication, and read-only verification.
- ### Claude Code
- Add a `.mcp.json` at the root of your own project (not this repo's root). Claude Code expands `${VAR}` from your environment, so the key stays out of the file:
- ```json
- {
- "mcpServers": {
- "harness-mcp-v2": {
- "command": "npx",
- "args": ["-y", "harness-mcp-v2"],
- "env": {
- "HARNESS_API_KEY": "${HARNESS_API_KEY}"
- }
- }
- }
- }
- ```
- `claude mcp add --scope project` writes the same file (see `claude mcp add --help`). An `mcpServers` entry in `~/.claude/settings.json` is not read as MCP configuration.
- ### Cursor
- Add to `~/.cursor/mcp.json` (user-level) or `.cursor/mcp.json` (project-level):
- ```json
- {
- "mcpServers": {
- "harness-mcp-v2": {
- "command": "npx",
- "args": ["-y", "harness-mcp-v2"],
- "env": {
- "HARNESS_API_KEY": "<your-api-key>"
- }
- }
- }
- }
- ```
- ### OpenAI Codex
- Codex reads MCP servers from `~/.codex/config.toml` as TOML tables, not JSON. `env_vars` forwards the named variables from your shell environment:
- ```toml
- [mcp_servers.harness-mcp-v2]
- command = "npx"
- args = ["-y", "harness-mcp-v2"]
- env_vars = ["HARNESS_API_KEY"]
- ```
- ### VS Code (GitHub Copilot)
- Add to `.vscode/mcp.json` in the workspace. This file is usually committed, so do not put a real key in it:
- ```json
- {
- "servers": {
- "harness-mcp-v2": {
- "command": "npx",
- "args": ["-y", "harness-mcp-v2"],
- "env": {
- "HARNESS_API_KEY": "<your-api-key>"
- }
- }
- }
- }
- ```
- ## Team Setup (Claude Code)
- To have teammates prompted to install the plugins, commit a project `.claude/settings.json`:
- ```json
- {
- "extraKnownMarketplaces": {
- "harness-skills": {
- "source": {
- "source": "github",
- "repo": "harness/harness-skills"
- }
- }
- },
- "enabledPlugins": {
- "harness-fme@harness-skills": true
- }
- }
- ```
- List one `enabledPlugins` entry per plugin the team uses (`<plugin>@harness-skills`).
- ## Operating Model
- The best Harness skills follow the same control flow even when they target different resource types:
- 1. **Establish scope first** - confirm account/org/project context before listing, creating, updating, or deleting resources.
- 2. **Verify dependencies before generating dependents** - do not reference connectors, secrets, environments, infrastructure, or templates that have not been confirmed to exist.
- 3. **Discover schema before writing payloads** - use MCP `harness_describe` or the installed CLI command specification and documented payload schema; stop when required capabilities are unavailable rather than guessing fields.
- These repo-level playbooks live in:
- - [`references/scope-establishment.md`](references/scope-establishment.md)
- - [`references/dependency-check-playbook.md`](references/dependency-check-playbook.md)
- - [`references/schema-validation-loop.md`](references/schema-validation-loop.md)
- - [`templates/operation-summary.md`](templates/operation-summary.md)
- ## Workflow Modes
- | Workflow mode | Representative skills | Use when |
- |---------------|-----------------------|----------|
- | Create and scaffold | `/create-pipeline`, `/create-service`, `/create-connector`, `/create-template`, `/fme-pipeline` | You need to define or generate new Harness resources and their YAML or MCP payloads. |
- | Run and debug | `/run-pipeline`, `/debug-pipeline`, `/migrate-pipeline`, `/manage-delegates` | You already have resources and need to execute, diagnose, or repair behavior. |
- | Govern and secure | `/manage-roles`, `/manage-users`, `/create-policy`, `/security-report`, `/audit-report` | You need RBAC, policy, compliance, or security workflows with blast-radius awareness. |
- | Analyze and report | `/dora-metrics`, `/analyze-costs`, `/scorecard-review`, `/template-usage` | You need structured reports, summaries, recommendations, or adoption analysis. |
- ## End-to-End Workflows
- ### New microservice setup
- Use these skills in order:
- 1. `/create-connector`
- 2. `/create-secret`
- 3. `/create-service`
- 4. `/create-environment`
- 5. `/create-infrastructure`
- 6. `/create-pipeline`
- 7. `/create-trigger`
- ### Debug a failed deployment
- Typical sequence:
- 1. `/run-pipeline` to identify the latest execution or reproduce the issue
- 2. `/debug-pipeline` to classify the failure and inspect root cause
- 3. `/template-usage` if shared templates may have propagated the issue
- 4. `/manage-delegates` if the failure points to delegate capacity or connectivity
- ## Migrating from the Flat Layout
- Skills moved from `skills/<skill>/` to `skills/<category>/<skill>/`; the categories are listed in `skills/categories.json`. Old deep links and `@harness-skills/skills/<skill>/SKILL.md` references need the category segment:
- ```
- Before: @harness-skills/skills/<skill>/SKILL.md
- After: @harness-skills/skills/<category>/<skill>/SKILL.md
- Example: @harness-skills/skills/pipelines/create-pipeline/SKILL.md
- ```
- Branches opened before the move: merge or rebase `main`, run `python3 scripts/migrate_to_categories.py --repair` (add `--map <new-skill>=<category>` for each skill the script does not know, repeatable), review the changes, and commit. The script is temporary and will be removed after the merge window.
- If you installed with `npx skills add`, re-run your install to pick up the new layout.
- ## Project Structure
- ```
- harness-skills/
- ├── skills/
- │ ├── categories.json # Categories and Claude Code plugins (source of truth)
- │ ├── pipelines/ # One folder per category
- │ │ ├── create-pipeline/
- │ │ │ ├── SKILL.md
- │ │ │ └── references/
- │ │ ├── debug-pipeline/
- │ │ │ └── SKILL.md
- │ │ └── ...
- │ ├── fme/
- │ └── ... # Nine categories in total
- ├── references/ # Shared repo-level playbooks
- ├── templates/ # Shared repo-level output templates
- ├── template/
- │ └── SKILL.md.template # Copy-ready skeleton for new skills
- ├── scripts/
- │ ├── validate-skills.sh # Structural validation of skills
- │ ├── generate_catalog.py # Regenerates manifests, skills.json and skill tables (--check for CI)
- │ ├── catalog_lib.py # Catalog loading shared by the generators
- │ ├── gen_claude.py # Claude Code marketplace generator
- │ ├── gen_cursor.py # Cursor plugin generator
- │ ├── gen_codex.py # Codex plugin generator
- │ ├── check_links.py # Markdown link and path-mention checker
- │ └── migrate_to_categories.py # Temporary flat-to-category migration helper
- ├── tests/ # Catalog, generator, manifest and link-checker tests
- ├── examples/
- │ ├── v0/ # v0 pipeline examples
- │ ├── v1/ # v1 pipeline examples
- │ ├── templates/ # Template examples
- │ ├── triggers/ # Trigger examples
- │ ├── services/ # Service definition examples
- │ ├── environments/ # Environment examples
- │ ├── connectors/ # Connector examples
- │ └── ...
- ├── .claude-plugin/marketplace.json # Generated: Claude Code marketplace (nine plugins)
- ├── .cursor-plugin/ # Generated: Cursor marketplace and plugin manifest
- ├── .agents/plugins/marketplace.json # Generated: Codex marketplace
- ├── .codex-plugin/plugin.json # Generated: Codex plugin manifest
- ├── skills.json # Generated: machine-readable skill index
- ├── .cursor/rules/harness.mdc # Project rules for Cursor
- ├── .github/
- │ ├── copilot-instructions.md
- │ └── workflows/validate-skills.yml # CI
- ├── AGENTS.md # Canonical project rules (Codex, Cursor, etc.)
- ├── CLAUDE.md # Imports AGENTS.md for Claude Code
- ├── CONTRIBUTING.md # Contribution guidelines
- ├── LICENSE
- └── README.md
- ```
- ## Shared Playbooks and Templates
- Use root-level shared assets for behavior that should stay consistent across many skills:
- - `references/` - reusable operating guidance such as scope establishment, dependency verification, and schema-validation loops
- - `templates/` - reusable output contracts such as operation summaries and report formats
- Use per-skill `references/` or `templates/` when the content is domain-specific and should not be imported broadly.
- ## Skill Anatomy
- Each skill is a directory at `skills/<category>/<skill>/` containing a `SKILL.md` with YAML frontmatter and a consistent markdown body: `# Title`, `## Instructions`, `## Examples`, `## Performance Notes`, and `## Troubleshooting`. Start from [`template/SKILL.md.template`](template/SKILL.md.template); [CONTRIBUTING.md](CONTRIBUTING.md) documents the frontmatter fields, description rules, and shared-reference conventions.
- For complex skills, add `references/` or `templates/` under the skill directory rather than bloating `SKILL.md`.
- ## MCP Tools
- MCP-powered skills use the [Harness MCP v2 server](https://github.com/harness/mcp-server), which provides 10 generic tools dispatched by `resource_type`:
- | Tool | Purpose |
- |------|---------|
- | `harness_list` | List resources |
- | `harness_get` | Get resource details |
- | `harness_create` | Create a resource |
- | `harness_update` | Update a resource |
- | `harness_delete` | Delete a resource |
- | `harness_execute` | Execute an action |
- | `harness_search` | Search across resources |
- | `harness_describe` | Get resource schema |
- | `harness_diagnose` | Diagnose issues |
- | `harness_status` | Check system status |
- ## Schema References
- - [v0 Pipeline/Template/Trigger Schema](https://github.com/harness/harness-schema/tree/main/v0)
- - [v1 Pipeline Spec](https://github.com/thisrohangupta/spec)
- - [Agent Templates](https://github.com/thisrohangupta/agents)
- - [Harness MCP v2 Server](https://github.com/harness/mcp-server)
- ## Contributing
- See [CONTRIBUTING.md](CONTRIBUTING.md) for skill authoring standards, validation rules, and contribution workflow. Before opening a PR, run the four checks it lists: `scripts/validate-skills.sh`, `python3 scripts/generate_catalog.py` (commit the generated files), `python3 scripts/check_links.py`, and `python3 -m unittest discover -s tests -v`.
- ## License
- This project is licensed under the [Apache License 2.0](LICENSE).
- Copyright 2026 Harness Inc.
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.