Accordo
by khaoss85
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_docsA | Keyword search across this framework's documentation set (README, AGENTS, ARCHITECTURE, PRODUCT, DECISIONS and everything under docs/). Returns the file path, the nearest heading and an excerpt for each hit. Read-only; it serves documentation, never customer records. |
| get_capabilityA | Resolve a capability from the claims ledger by id (C-nn), by standing-limitation id (L-nn), or by topic. Every capability is returned together with the evidence that proves it and the limitation that bounds it — this tool cannot return one without the other. |
| check_jobA | Answer "can this framework do X?" against the CRM jobs-to-be-done index. Returns the matching jobs with their status (not supported / partially supported / technically supported / validated end to end), the tests that prove them, and an explicit answer when the truth is that the job is not supported. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| documentation-index | Every document this server serves, what it is, and what is deliberately excluded. |
| claims-ledger | Every public capability claim bound to the tests that prove it, each paired with the limitation that bounds it, plus the standing limitations of the framework. |
| jobs-index | The CRM jobs index generated from docs/benchmarks/CRM_JTBD_MATRIX.md, each job carrying one of four statuses. Absent from a checkout until `node scripts/generate-jobs.js` has been run. |
| AGENTS.md | This repository is an agent-native CRM framework. Preserve the separation between deterministic CRM state and AI-assisted development/orchestration. |
| ARCHITECTURE.md | CRM state is deterministic. AI agents interpret the user's intent and compose the system, but state changes pass through services and workflows. |
| CLAUDE.md | Read `AGENTS.md` first and follow it as the canonical repository guidance. |
| CONTRIBUTING.md | Thank you for looking. This project has an unusual amount of written-down process for its size, |
| DECISIONS.md | **Status:** accepted |
| GEMINI.md | Read `AGENTS.md` first and follow it as the canonical repository guidance. It is |
| PRODUCT.md | Accordo is a framework that lets a user describe a commercial process to Codex or Claude Code and have the agent safely build, run, inspect and improve a CRM a… |
| README.md | **Describe your sales process to your coding agent; own the CRM it builds.** |
| ROADMAP.md | This file is a **pointer, not a ledger.** It used to carry a four-milestone plan |
| SECURITY.md | This framework is **local-development-only** and has no production security spine. That is |
| TASKS.md | The first unchecked item is the default next task for Codex. |
| docs/ACTIONS.md | A **record action** is a lifecycle operation on one record that is more than a |
| docs/ADMIN.md | The Admin is a zero-build, dependency-free static app (`apps/admin/public/`) served by the HTTP server at `/`. It has two parts: the handwritten CRM dashboard … |
| docs/ADMIN_SMOKE.md | Automated tests cover the Admin at the DOM/integration level (real server + real fetch + a fake document) and run in CI. **They do not drive a real browser, an… |
| docs/AGENT_HARNESS_COMPATIBILITY.md | Accordo is not built for one coding agent. AX1 (`app inspect`) and AX2 |
| docs/API.md | Base URL: `http://localhost:4000` |
| docs/APPLICATION_INSPECTION.md | One deterministic document describing the application a project has actually |
| docs/CODER_TOOLING_ROADMAP.md | The surfaces a coding agent uses to understand, plan and change an Accordo |
| docs/COMMERCIAL_OPERATIONS.md | Catalog, quotes and discount approval over the B2B starter — local development |
| docs/CONTRACT_ACTIVATION.md | Turning a signed immutable Order into a live commercial state: one Commercial |
| docs/DELIVERY_CHANGE_ACCEPTANCE.md | M14b1 recorded what a delivery project **consumed**. M14b2 records what changed |
| docs/DELIVERY_ECONOMICS.md | M13 planned a delivery project. M14a let a human run it. M14b1 records **what |
| docs/DELIVERY_HANDOVER.md | Turning the pending Delivery Obligations of an activated contract into a |
| docs/FALSIFY.md | node scripts/falsify.js # the fast set — a couple of seconds |
| docs/HANDOVER_CODEX.md | Target GitHub owner: `khaoss85` |
| docs/IMPLEMENTATION_EVIDENCE.md | `accordo solution verify <plan.json> --evidence <evidence.json> [--json] [--root dir]` — **DX10**. |
| docs/JTBD.md | “I need this commercial outcome” |
| docs/LEAD_INTELLIGENCE.md | Enrichment, explainable scoring and deterministic routing over the Lead |
| docs/MCP.md | The server uses newline-delimited JSON-RPC over stdio. Diagnostics are written only to stderr. |
| docs/MODULE_EVOLUTION.md | How a record that already exists — with rows in it — gains a field, a status or |
| docs/MODULE_FACTORY.md | The module factory turns a validated manifest (`docs/MODULE_MANIFEST.md`) into a complete, runnable backend module: readable service code, deterministic migrat… |
| docs/MODULE_MANIFEST.md | A module manifest is a small declarative JSON file that describes one CRM module entity. The framework validates it with precise errors and generates determini… |
| docs/PACKAGE_AUTHORING.md | How Claude Code, Codex — or a person — adds a bounded domain to an Accordo |
| docs/PROJECT_STATUS.md | The single operational snapshot of the repository. Volatile facts — merged |
| docs/QUALITY_GATES.md | The review discipline that produced Milestones 6–11 written down as repository |
| docs/RENAME_SURFACE.md | The complete inventory of where the working title lives, grouped by what breaks when it moves. |
| docs/RENEWAL_AMENDMENT.md | **Milestone M16b. ADR-035. Packages: `contracts` (v7) and `lifecycle` (v2).** |
| docs/SCENARIO_EVIDENCE.md | `accordo scenario run <scenario> [--json] [--root dir]` — **DX6**. |
| docs/SERVICE_OPERATIONS.md | The **third** domain package built under ADR-018, and the second that depends on |
| docs/SIGNATURE_ORDER.md | Turning an approved Quote Version into a signed commitment and one immutable |
| docs/SKILL_PACKAGING.md | A skill is the part of this framework that installs into somebody else's agent. |
| docs/SOLUTION_PLAN.md | AX1 answers *what has this project actually composed*. This answers the next |
| docs/WORK_TASKS.md | **Work v1 (ADR-030), the `work` package.** Follow-up work a person must do, and |
| docs/architecture/AGENT_TOOL_SURFACE.md | **Status: strategy and proposed policy. Nothing here is implemented.** No tool |
| docs/architecture/EXTRACTION_PREPARATION.md | **Nothing in this document is implemented, and nothing in it authorizes an |
| docs/architecture/INTELLIGENCE_PACKAGE_TARGET.md | **Status: accepted target, nothing implemented.** ADR-021 and ADR-022 are |
| docs/architecture/LEGACY_ALIGNMENT_MATRIX.md | **Status: assessment and policy. No domain is refactored by the PR that |
| docs/benchmarks/AGENT_TOOL_SELECTION_PROTOCOL.md | **Status: protocol and pilot instrument. No product command, MCP tool, runtime or |
| docs/benchmarks/CRM_JTBD_MATRIX.md | A validation catalogue of classic CRM jobs-to-be-done, tracking what the framework genuinely supports today. This is **not** a marketing claim: a JTBD is marke… |
| docs/benchmarks/PILOT_PROTOCOL.md | How one operator runs one benchmark prompt, end to end, and what they are allowed |
| docs/benchmarks/TOOL_SELECTION_PILOT_2026-08-13.md | Date: 2026-08-13 |
| docs/benchmarks/URR_PILOT_2026-08-10.md | Date: 2026-08-10 |
| docs/evidence/commercial-extraction.md | The second legacy domain to leave `packages/core`, on the pattern the Lead |
| docs/evidence/lead-intelligence-extraction.md | **What this is.** The rails an agent actually used to move a legacy domain into |
| docs/marketing/AWESOME_LIST_SUBMISSIONS.md | Two lists, two different mechanisms, and neither is a PR an agent can open from |
| docs/marketing/BENCHMARK_PUBLICATION.md | What may be said in public about an Edition L run, and what may not. This is a |
| docs/marketing/CONTENT_PILLARS.md | The organic-growth engine in `docs/strategy/ORGANIC_GROWTH.md` defines the loop and its six |
| docs/marketing/CONTENT_PRODUCTION.md | `CONTENT_PILLARS.md` says **what** we write and in what order. `ORGANIC_GROWTH.md` |
| docs/marketing/CORRECTIONS.md | Every public claim this project made and then had to withdraw or narrow, with the commit that |
| docs/marketing/DEPLOYMENT.md | How `accordo.dev` gets served, what is already configured, and the two things a person has to |
| docs/marketing/DESIGN_BRIEF.md | Hand this to a designer who has no access to the repository. Everything needed |
| docs/marketing/FOUNDER_CHECKLIST.md | Everything an agent cannot do, in the order that matters, with the exact commands and clicks. |
| docs/marketing/GITHUB_LISTING.md | Everything the repository's public surface needs, written out so applying it is mechanical. |
| docs/marketing/LAUNCH_PACKET.md | Copy for the two launch channels, written now so that it is written carefully rather |
| docs/marketing/NAME_VERIFICATION.md | Re-verification of the `BRAND_REQUIREMENTS.md` shortlist, so the founder's remaining work is a |
| docs/marketing/OBJECTIONS.md | The fifteen hardest things a sceptical developer says, and the answer in the project's own |
| docs/marketing/PENDING_HUMAN_SUBMISSION.md | Everything an agent has prepared and **must not** send, create or register, and the person-only |
| docs/marketing/SITE_ARCHITECTURE.md | The information architecture of the public site, and the contract every generated |
| docs/marketing/drafts/2026-08-09-indie-hackers-refusal-surface.md | I am building Accordo, an open-source framework that coding agents use to build |
| docs/strategy/AGENT_CRM_CLOUD.md | **Status: product design and roadmap track only. Nothing in this document is implemented.** No control plane, no managed runtime, no Cloud CLI, no project MCP … |
| docs/strategy/AGENT_DISCOVERY.md | How this framework becomes discoverable — and, where possible, recommendable — in the channels where coding agents and their users decide what to build with. R… |
| docs/strategy/AGENT_RECOMMENDATION.md | `AGENT_DISCOVERY.md` answers *which channels exist and how they work*. This document answers a |
| docs/strategy/ANALYTICS_STUDIO.md | **Status: product strategy and roadmap only. Nothing in this document is implemented.** No semantic model, metric definition, dataset, report, dashboard, widge… |
| docs/strategy/BRAND_REQUIREMENTS.md | The current name is a working title, and it fails as a public brand on every axis that matters for this strategy: |
| docs/strategy/CAMPAIGNS_JOURNEYS.md | **Status: product strategy only. Nothing in this document is implemented.** No campaign, proposal, audience, suppression, journey, enrolment, content asset, la… |
| docs/strategy/CATEGORY.md | **The agent-native CRM framework**: the open-source framework that coding agents (Claude Code, Codex) use to *generate* bespoke CRM applications as code the cu… |
| docs/strategy/CLOUD_JTBD.md | Operator and builder jobs for the optional managed layer. `AGENT_CRM_CLOUD.md` |
| docs/strategy/CODING_AGENT_DX_NORTH_STAR.md | **Status: canonical decision rule.** This document does not describe a feature. |
| docs/strategy/COMPETITOR_MAP.md | Research date: August 4, 2026. GitHub figures read from repo pages on that date (rounded as GitHub displays them). Items marked *(unverified)* came from second… |
| docs/strategy/CONTRACT_SUBSCRIPTION_RENEWAL.md | **Status: design only. None of the primitives below exists.** |
| docs/strategy/CRM_BUILD_BENCHMARK.md | A public, reproducible benchmark measuring whether a coding agent can turn a CRM brief into a working, deployed CRM using this framework — and how it compares … |
| docs/strategy/CUSTOMER_REVENUE_OS_ROADMAP.md | **Status: strategy. Nothing here is implemented, and nothing here renames |
| docs/strategy/DATA_GOVERNANCE.md | **Status: design only. None of this is implemented, and nothing here is legal |
| docs/strategy/DELIVERY_SERVICE.md | **Status: product strategy and roadmap only. Nothing in this document is implemented.** No delivery project, milestone, partner engagement, time tracking, chan… |
| docs/strategy/DESIGN_TO_CRM.md | **Status: design only. The generated Admin exists; none of the design pipeline |
| docs/strategy/DISTRIBUTION_SUBMISSIONS.md | Every distribution action, in dependency order, with the exact command or the |
| docs/strategy/EXECUTION_ROADMAP.md | Phased path from the current vertical slice to the public, agent-recommended CRM framework. Every phase lists: outcome, deliverables, dependencies, acceptance … |
| docs/strategy/EXPERIMENTATION_ATTRIBUTION.md | **Status: product strategy only. Nothing in this document is implemented.** No experiment, variant, control group, holdout, assignment, media plan, funnel defi… |
| docs/strategy/EXTERNAL_REVIEW.md | An outside assessment of how well this framework serves a coding agent (Claude Code, |
| docs/strategy/GO_TO_MARKET.md | How this framework reaches the people who should use it, what is built, what is prepared but |
| docs/strategy/GTM_TECHNICAL_EVIDENCE_HANDOFF.md | **This is not marketing copy, and it does not replace the GTM claims ledger.** |
| docs/strategy/INTEGRATION_RUNTIME.md | **Status: design only. Nothing in the "missing" half of this document exists.** |
| docs/strategy/JOBS_AND_OUTBOX.md | **Status: design only. Nothing here is implemented.** |
| docs/strategy/MARKETING_GROWTH_OPERATIONS.md | **Status: product strategy and roadmap only. Nothing in this document is implemented.** No campaign, audience, journey, experiment, content asset, landing page… |
| docs/strategy/MASTER_PLAN.md | The canonical entry point to the strategy. Read this first; follow the links for depth. For implementation work, `ARCHITECTURE.md` and `DECISIONS.md` remain th… |
| docs/strategy/MEDUSA_PLAYBOOK.md | What Medusa did to become the default open-source commerce framework for JavaScript developers, what of it transfers to an agent-native CRM framework, and what… |
| docs/strategy/NORTH_STAR_EXPERIENCE.md | A user opens Claude Code or Codex and provides three inputs: a CRM brief in natural language, a description of one or more business processes, and a design ref… |
| docs/strategy/OBJECTIVE_DRIVEN_AGENT_EXPERIENCE.md | **Status: North Star and design. The end-to-end experience is not implemented.** Parts of it work today — package and capability discovery, custom package auth… |
| docs/strategy/OBJECTIVE_DRIVEN_FUNNEL_EXAMPLE.md | **Status: a worked design, not a transcript of something that ran.** Parts of the solution below are buildable today; the analytics and campaign parts are not,… |
| docs/strategy/ORGANIC_GROWTH.md | Growth built only on real product assets: working code, reproducible builds, honest benchmarks. No paid acquisition, no inflated claims, no content that a main… |
| docs/strategy/PLATFORM_ALIGNMENT_GATE.md | A short architecture and roadmap checkpoint taken **after Milestone 11 merged |
| docs/strategy/PLATFORM_CAPABILITIES.md | One bounded vocabulary shared by every kind of module: **handwritten core |
| docs/strategy/RECOMMENDATION_MAP.md | The complete map from **what a user asks their coding agent** to **the artifact |
| docs/strategy/REVENUE_OPERATIONS.md | **Status: product strategy and roadmap only. Nothing in this document is implemented.** No enrichment provider, scoring model, routing policy, product catalog,… |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/khaoss85/agent-crm'
If you have feedback or need assistance with the MCP directory API, please join our Discord server