Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/lennix1337/genexus18mcp/gemini-mdgit clone --depth 1 https://github.com/lennix1337/Genexus18MCPWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01802 | $0.01802 |
| Opus 5 | $0.00901 | $0.00901 |
| Sonnet 5 | $0.00360 | $0.00360 |
| Haiku 4.5 | $0.00180 | $0.00180 |
Grade A, and why
Genexus18MCP GEMINI.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GeneXus MCP Protocol Guide
Required repo skills before editing KB logic:
This repository is MCP-first. The official transport is MCP over stdio or HTTP at /mcp. The old /api/command endpoint is no longer part of the gateway surface.
Correct MCP flow
- Initialize the session with
initialize. - Discover the live surface with
tools/list,resources/list,resources/templates/list,prompts/list, andcompletion/completewhen needed. - Execute work with
tools/call,resources/read, andprompts/get. - For HTTP MCP, send
MCP-Protocol-Version: 2025-06-18and reuse the returnedMCP-Session-Id.
Recommended tool usage
genexus_query: find objects, references, signatures, and dependency entry points. Supports optionaltypeFilteranddomainFilterfor server-side narrowing.genexus_read: read object parts with pagination. Pass singularnamefor one object ortargets[]for coordinated multi-object reads (mutually exclusive). For MCP clients, keep reads paginated; the server intentionally returns a source-first first page when no explicitoffset/limitis provided. For XML metadata parts such asLayout,WebForm, andPatternInstance, the gateway applies a larger metadata budget to avoid truncating the editable XML.genexus_edit: apply focused edits to a part or replace content through the MCP write path. Three modes:mode: 'xml'(default): full XML replacement.mode: 'ops': typed semantic op catalog (set_attribute,add_attribute,remove_attribute,add_rule,remove_rule,set_property).mode: 'patch': JSON-Patch (RFC 6902) array over the canonical JSON object representation; legacy string-form text patch is also accepted for backward compatibility. Passtargets[](array of edit-request objects) for atomic multi-object edits — mutually exclusive with singularname.PatternInstancetargets the authoritative WorkWithPlus instance when the requested object is a WebPanel managed by WorkWithPlus.
- All write tools (
genexus_edit,genexus_create_object,genexus_refactor,genexus_forge,genexus_import_object) accept:dryRun: true— returns a preview envelope{_meta:{dryRun,schemaVersion}, plan:{touchedObjects, xmlDiff, brokenRefs, warnings}}without mutating the KB.idempotencyKey: '<token>'([A-Za-z0-9_-]{1,128}) — deduplicates retries; concurrent calls with the same key are coalesced; cache is per-KB, sliding TTL 15 min default.dryRunbypasses the cache.
genexus_inspect: get structured conversion or object context.genexus_analyze: one tool withmode:linter | navigation | hierarchy | impact | data_context | ui_context | pattern_metadata | summary | explain(the last two replaced the standalonegenexus_summarizeandgenexus_explain_codein v2.3.0;explaintakescode).genexus_lifecycle: build, validate, index, and KB lifecycle operations.genexus_sql:action=ddlfor Transaction/Table DDL oraction=navigationfor the SQL of a procedure's For Each (replacesgenexus_get_sqlandgenexus_get_sql_for_navigationin v2.3.0).genexus_kb: multi-KB pool management (action:list | open | close | set_default). Replacesgenexus_open_kbin v2.3.0.genexus_create_object: create new KB objects.genexus_refactor: supported rename and extraction refactors.genexus_add_variable: add variables through the worker contract.genexus_format: format source through the worker formatter.genexus_properties: get or set object properties.genexus_asset: find, read, and write KB assets such as.xlsxtemplates. Reads are metadata-first; requestincludeContent=trueonly for files small enough to carry as Base64 safely.genexus_history: list, read, save, and restore object history.genexus_structure: read or update logical and visual structure.genexus_doc: access documentation, health, and visualization flows.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 86 lines · 1,802 tokens per session scan A ad75c24dd0c0
Genexus18MCP GEMINI.md is an instructions file published in the GitHub repository lennix1337/Genexus18MCP (28 stars, last pushed 4d ago), licensed MIT. It adds 1,802 tokens to every session, about $0.0090 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
ai-workflow AGENTS.md
Instructions for cunhaax/ai-workflow, covering ai workflow template, rules — non-negotiable, project overview, commands and architecture.
trackly-cli CLAUDE.md
Claude Code instructions for trackly-app/trackly-cli, covering trackly-cli, tech stack, backend production source of truth, directory structure and key commands.
com-example-config-AsyncTestConfig
Instructions for PIsberg/vibetags, covering copilot instructions for asynctestconfig and immutable type.
pupila CLAUDE.md
Instructions for FranRom/pupila, covering claude.md, where things live, stack, run locally and repo layout.
claude-bridge AGENTS.md
Instructions for kadaba/claude-bridge, covering agents.md — using claude-bridge from any ai coding cli, two independent channels, 1. coordination (the conversation), 2. file transfer (the bytes) — same for every tool and workflows.
UsageAtlas AGENTS.md
Instructions for SRX9/UsageAtlas, covering repository guidelines and coding approach.