{
  "markdown": "<p align=\"center\">\n  <img src=\"./assets/logo.svg\" alt=\"CRM Workflows — 36 evidence-first skills for safer revenue work\" width=\"900\">\n</p>\n\n<p align=\"center\">\n  <a href=\"https://github.com/sepivip/crm-workflows/actions/workflows/validate.yml\"><img alt=\"Validation\" src=\"https://github.com/sepivip/crm-workflows/actions/workflows/validate.yml/badge.svg\"></a>\n  <a href=\"https://github.com/sepivip/crm-workflows/releases/tag/v0.1.0\"><img alt=\"Release\" src=\"https://img.shields.io/github/v/release/sepivip/crm-workflows?color=6C5CE7\"></a>\n  <a href=\"https://agentskills.io\"><img alt=\"Agent Skills\" src=\"https://img.shields.io/badge/standard-Agent%20Skills-17152F\"></a>\n  <a href=\"./LICENSE\"><img alt=\"MIT License\" src=\"https://img.shields.io/badge/license-MIT-54BFA9\"></a>\n  <img alt=\"36 skills\" src=\"https://img.shields.io/badge/skills-36-6C5CE7\">\n</p>\n\n<p align=\"center\"><strong>Portable CRM judgment for revenue agents—without hard-coding a CRM vendor or silently taking action.</strong></p>\n\nCRM Workflows is a public collection of 36 sales and revenue-operations skills built on the open [Agent Skills specification](https://agentskills.io). The same core instructions can be packaged for Codex and ChatGPT, Claude Code, Gemini CLI, GitHub Copilot, or another compatible agent.\n\nThe workflows are intentionally evidence-first. They distinguish facts from inference, preserve ambiguity, cite consequential claims, minimize sensitive-data access, and require review before writes, sends, scheduling, assignments, or other external side effects.\n\nThe shared [CRM-neutral semantic model](./references/crm-neutral-model.md) explains how provider-specific records map into those workflows.\n\n## What you get\n\n| Area | Skills | Examples |\n| --- | ---: | --- |\n| Plan and pipeline intelligence | 8 | Daily briefing, pipeline review, forecast narrative, slip scenarios |\n| Deal strategy and execution | 6 | Deal review, stakeholder map, close plan, opportunity updates |\n| Account growth and prospecting | 10 | Account plan, lead triage, renewal radar, expansion whitespace |\n| Meeting and communication productivity | 6 | Call prep, transcript summary, inbox sweep, follow-up |\n| Governance, activity, and data hygiene | 6 | Win/loss review, activity history, internal answers, hygiene audit |\n\nBrowse the complete [36-skill catalog](./docs/CATALOG.md).\n\n## Install\n\n### OpenAI Codex and ChatGPT\n\n```bash\ncodex plugin marketplace add sepivip/crm-workflows\ncodex plugin add crm-workflows@crm-workflows\n```\n\nRestart the host if it asks you to refresh installed plugins.\n\n### Claude Code\n\n```bash\nclaude plugin marketplace add sepivip/crm-workflows\nclaude plugin install crm-workflows@crm-workflows\n```\n\nFor local development, clone the repository and run `claude --plugin-dir ./crm-workflows`.\n\n### Gemini CLI\n\n```bash\ngemini extensions install https://github.com/sepivip/crm-workflows\n```\n\n### GitHub Copilot or another Agent Skills host\n\nClone the repository, then copy all 36 skills—or only the ones you want—into a discovery directory supported by your agent:\n\n```bash\npython scripts/install_skills.py --target .agents/skills\npython scripts/install_skills.py --target .agents/skills --skill call-prep --skill deal-review\n```\n\nThe installer refuses to overwrite an existing skill directory. See the [compatibility guide](./docs/COMPATIBILITY.md) for personal-scope locations and invocation details.\n\n## Use\n\nAsk naturally when your host supports automatic skill selection:\n\n```text\nBrief me for tomorrow's Acme security review.\nWhat materially changed across my open deals this week?\nTurn these call notes into a reviewed follow-up package.\nAudit this opportunity export for hygiene issues—but do not change anything.\n```\n\nOr invoke a workflow explicitly when you want deterministic selection:\n\n```text\nUse $call-prep for tomorrow's Acme security review.\nUse $deal-review on the Globex renewal.\nUse $crm-hygiene-check on this export and remain read-only.\n```\n\nClaude Code exposes installed plugin skills as `/crm-workflows:call-prep`, `/crm-workflows:deal-review`, and so on. Invocation syntax varies by host; the skill logic does not.\n\n## The operating contract\n\nEvery workflow follows the same safety model:\n\n```text\nScope → Resolve identity → Read minimum data → Normalize semantics\n      → Separate evidence / inference / conflict / unknown\n      → Draft a reviewable result → Confirm → Act once → Report exactly\n```\n\n- CRM-neutral semantics: organization, person, lead, opportunity, activity, task, owner, stage, value, close date, and forecast category.\n- Evidence over confidence theater: unsupported probabilities and invented facts are prohibited.\n- Least-privilege retrieval: private or sensitive data is excluded unless necessary and authorized; sensitive personal data is never used for prospecting.\n- Review-gated mutations: proposals show the target, field-level diff, evidence, and side effects before execution.\n- Honest failure handling: no silent retries, no fabricated success, and no more than one bounded retry for a failed write.\n\nThe repo does **not** bundle a CRM connector or credentials. An agent can analyze supplied exports immediately; live retrieval and mutations require tools that the host exposes and the user authorizes.\n\n## Quality and evaluations\n\nThe repository ships 119 generated evaluation cases and a fictional CRM fixture:\n\n- 36 behavioral cases—one per skill\n- 36 implicit-routing cases\n- 36 incomplete-input cases\n- 8 high-risk safety cases\n- 3 unrelated non-trigger cases\n\nRun the full structural suite:\n\n```bash\npython -m pip install -r requirements-dev.txt\npython scripts/generate_skills.py --check\npython scripts/validate_crm_plugin.py\npython scripts/run_evals.py\n```\n\nExport self-contained JSONL cases for your preferred agent harness:\n\n```bash\npython scripts/run_evals.py --group behavioral --skill deal-review --emit-jsonl\n```\n\nThe validator checks the official Agent Skills schema, all provider manifests, generated-file drift, UI metadata, privacy boundaries, fixture references, and complete evaluation coverage. Behavioral model outputs still require an external agent harness or human grading; the repository does not claim model-independent semantic pass rates.\n\n## Build from the source catalog\n\n`scripts/generate_skills.py` is the reviewed source of truth for skill content and generated tests:\n\n```bash\npython scripts/generate_skills.py\n```\n\nThis regenerates every `SKILL.md`, every OpenAI UI manifest, the evaluation catalog, and the human-readable skill catalog. Read [CONTRIBUTING.md](./CONTRIBUTING.md) before proposing a workflow change.\n\n## Project map\n\n```text\nskills/                         36 portable Agent Skills\n  <skill>/SKILL.md              provider-neutral workflow\n  <skill>/agents/openai.yaml    optional Codex/ChatGPT UI metadata\nevals/                          routing, behavior, safety, and non-trigger cases\nreferences/                     shared CRM-neutral semantic model\nscripts/                        generator, validator, evaluator, installer\n.codex-plugin/                  OpenAI plugin manifest\n.claude-plugin/                 Claude plugin and marketplace manifests\ngemini-extension.json           Gemini CLI extension manifest\n```\n\n## Contributing and security\n\nIssues and pull requests are welcome. Start with [CONTRIBUTING.md](./CONTRIBUTING.md); report vulnerabilities privately according to [SECURITY.md](./SECURITY.md).\n\nReleased under the [MIT License](./LICENSE).\n",
  "bytes": 7458,
  "sha": "40e744423f59e543b871ea27244e0238ac8b692316fa6f7240d70566ed1758b1",
  "repo_slug": "sepivip/crm-workflows",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_sepivip_crm_workflows_c8ec6e02/readme"
}