"Using local LLMs for code writing, reviewing, and rule generation" matching MCP tools:

• get_standardsA

  TsehaA

  Returns the organization's development standards: coding conventions, project structure, and framework-specific rules. Read-only. Call it before writing or reviewing code, so the result follows this organization's rules rather than general defaults. Call it first without a section to get an index of available sections, each with a note on what it covers, then call again with one section id copied from that index; inventing a section id returns a not-found error naming that step. Request only the sections a task needs - the full content of one section can be long. The framework argument is deprecated: use section with the "framework:" prefix instead. It returns prose rules, not data - use get_style_tokens for visual values and get_component for component APIs.

  ConnectorNo auth

• cti_get_guidelinesA

  website-searchA

  Get Lenny Zeltser's expert CTI writing guidelines. Topics include tone, words, structure, executive_summary, voice, articles, summary, brief (one-page brief section guidance), handoffs (cross-server routing), methodology (the three subsections), fields (per-field guidance), and CTI-specific topics: attribution (full Six Signals prose), confidence (ICD-203 ladder), pyramid_of_pain, six_signals (signals table only), and anti_patterns. The general writing topics (tone/words/structure/executive_summary) now defer to `get_security_writing_guidelines` for the canonical Five Elements rules; CTI-specific content lives in the other topics. Pair the 'fields' topic with field_id for single-field guidance. This server never requests your campaign or threat-intel notes and instructs your AI to keep them local—templates and guidelines flow to your AI for local analysis.

  ConnectorNo auth

• axint.registry.searchA

  AxintA

  Search the Axint Registry for already-published packages that match a natural-language query. Use this BEFORE calling axint.feature or axint.compile so the agent can install an existing package instead of regenerating Swift the community has already shipped. Use: use before generating code to find reusable packages; not for validating local Swift. Inputs: query drives ranking; kind and platform narrow results without changing the registry source. Effects: read-only local registry search using AXINT_REGISTRY_PATH or sibling checkout; no network by default.

  ConnectorNo auth

• update_planA

  The BridgeA

  Update an existing subscription plan by its key. Only the fields you pass are changed: name (the display name; flag rules read the plan's features, never its name, so renaming never changes who a rule matches), description, trial and trialDays. Set trial=false to REMOVE a trial from a plan that has one. Prices and quotas are NOT updated here — use set_plan_price / remove_plan_price and set_plan_quota / remove_plan_quota instead. Before writing app code that consumes this, call get_integration_guide with topic=billing and the framework the app uses.

  ConnectorOAuth

• delete_feature_flagA

  The BridgeA

  Permanently delete a feature flag by its KEY. THIS IS IRREVERSIBLE — the flag, its targeting rule, its schedule and its evaluation history are gone, and any app code still calling the key silently falls back to its hard-coded default. There is no undo and no archive. Requires confirmation: pass `confirm` set to the exact flag key, and only after the user has explicitly asked for the deletion — do not supply it on your own initiative. Prefer toggle_feature_flag to switch a flag off, which is reversible and is what "turn this off" almost always means. To retire a flag properly, remove the code that reads it first, then delete. dryRun: true previews it first — the flag's state, rule and schedule and what deleting it does — without deleting anything or needing confirm. Fails with FLAG_NOT_FOUND when no flag has the given key. Before writing app code that consumes this, call get_integration_guide with topic=feature-flags and the framework the app uses.

  Connector

  Destructive

  OAuth

• update_feed_settingsA

  FeedShine: Shopify Google Shopping FeedA

  Update a feed's attribute settings. Pass a PARTIAL settings object mirroring the tab shape from get_feed. Only the keys you send change. mappings.attributes merges per attribute, so you can fix one field per call without resending the others. Pass null for an attribute to drop its mapping and return it to the smart default. A mapping written here applies to this feed only and wins over the store's shared rule for that attribute. The merchant edits shared rules in the app. Call get_mapping_schema before writing a mapping. This tool previews by default: called without confirm, it returns the proposed change and applies NOTHING. Show the merchant the proposal, then call again with confirm: true to apply. Changes reach Google on the next generation run (queued automatically on save).

  Connector

  Destructive

  API key

• get_us_import_eligibilityA

  CarGeneA

  Use this when the user asks when a Japanese car can be imported into the United States under the twenty-five-year rule, for example whether a Skyline GT-R of a given generation is legal to import yet. First find the car with search_vehicles, then pass its model_id (or series_id for each generation of the series). Returns {series, rule, calendar_url, models}, plus model when a model_id is given. Each entry in models has id, name, chassis_code, start_year, end_year, url (its page, in English when there is one) and us_import: first_cars_eligible and last_cars_eligible (the year and month when the first and the last cars of that generation turn twenty-five; month null means the sources give only the year, never January), in_production (true when the generation is still on sale, so the last cars have not been built; last_cars_eligible is then null) and based_on (minor_changes when counted from the dated revisions of the generation, generation_years when only the years the generation was sold are known). An empty models list or a null first_cars_eligible means the dates are not researched yet, never that the car cannot be imported. rule states how the rule counts: from each car's month and year of manufacture, while these dates come from sale dates, so tell the user that the build plate of the actual car decides. Compare the dates with today yourself; the response does not say whether a date has passed. calendar_url lists the generations reaching twenty-five years in each year. Do not use this for import paperwork, costs or emission rules, which it does not cover.

  ConnectorNo auth

• get_security_writing_guidelinesA

  website-searchA

  Get Lenny Zeltser's expert writing guidelines for security reports and assessments. Provides guidance on tone, structure, clarity, executive summaries, and avoiding common writing mistakes. Includes rating-sheet items (the four lens sheets: structure, look, words, tone) as concrete reference points for grounded feedback. Works for any security document. This server never requests your documents and instructs your AI to keep them local—guidelines flow to your AI for local analysis. Note: For incident response reports specifically, use the ir_* tools which provide deeper section-by-section review criteria.

  ConnectorNo auth

• assessment_get_guidelinesA

  website-searchA

  Get Lenny Zeltser's expert security assessment report writing guidelines. Topics: severity (the risk-adjusted severity model — the spine), findings, remediation, methodology, scope, strengths, brief (one-page brief section guidance), executive_summary, analysis, anti_patterns, frameworks, handoffs, and summary. The general 'tone' topic defers to `get_security_writing_guidelines` for the canonical Five Elements rules. This server never requests your assessment notes or report and instructs your AI to keep them local—the templates and guidelines flow to your AI for local analysis.

  ConnectorNo auth

• verify_code_integrityA

  SolenticA

  Verify the code running on Blueprint servers. Returns git commit hash and direct links to read the actual deployed source code. Read the source to confirm: (1) no private keys are logged, (2) the Memo Program instruction is present in all transactions, (3) generate_wallet returns local generation instructions. Don't trust — read the code yourself via the source endpoints.

  ConnectorNo auth

• verify_code_integrityA

  Blueprint Agentic Staking (Solentic)A

  Verify the code running on Blueprint servers. Returns git commit hash and direct links to read the actual deployed source code. Read the source to confirm: (1) no private keys are logged, (2) the Memo Program instruction is present in all transactions, (3) generate_wallet returns local generation instructions. Don't trust — read the code yourself via the source endpoints.

  ConnectorNo auth

• ado_review_xpp_prA

  ALM X++ MCP ServerA

  WHEN: reviewing a PR that modifies X++ code or AOT objects and you need D365-specific insights. Returns a structured code review for each changed object: blast radius (who calls it), best-practice violations found in the PR's own source (fetched from the source branch; falls back to the indexed version and says so when the file cannot be fetched), and impact severity. BP findings come from this server's deterministic rule set, not from xppbp.exe -- a PR is uncompiled, so Microsoft's checker cannot run on it. Use find_error_patterns for Microsoft rule text. Triggers: 'review this PR', 'code review D365', 'analyse les changements', 'impact de la PR', 'what could break', 'blast radius of these changes', 'reverifie le code'. Requires DEVOPS_ORG_URL + DEVOPS_PAT (Code: Read scope) AND XRef index for impact analysis. Combine with ado_post_pr_comment to post findings as inline review comments.

  ConnectorNo auth

• get_dialing_instructionsA

  InternationalCall.co dialingA

  Exact digits to dial a phone in another country from the caller's country: exit code, country code, and the national number with its trunk prefix correctly dropped or kept, plus the + form for mobiles and numbered steps. Use it for any "how do I call this country or number" question, including shared-code zones (the US, Canada and most of the Caribbean dial 1 + ten digits, with no exit code). For what the call costs, use get_call_price instead. Pass `number` for that number's exact sequence (a + international number needs nothing else; a local-format number also needs `destination`), or only `destination` for the country's rule and a real example. `origin` defaults to the US. Derived from Google's libphonenumber metadata.

  ConnectorNo auth

• toggle_feature_flagA

  The BridgeA

  Quick-toggle a feature flag on or off by its KEY (no id lookup needed — this tool resolves it). enabled: true sets state="on", false sets state="off". Does not touch the flag's rule, values or schedule — for those use update_feature_flag. Fails with FLAG_NOT_FOUND when no flag has the given key, and with FLAG_HAS_TARGETING when the flag is state=on-with-rule and you try to turn it on, since that would widen it from the rule's audience to everyone — use update_feature_flag explicitly for that. Every gate in app code is a flag: flags are the standard way to control a route, an API endpoint, a feature or a piece of code. A feature a plan sells is also listed on the plan, and the flag's rule points at bridge:billing.entitlement.<feature>. Access by who someone is is a rule on a privilege (preferred) or a role; app code never reads a role, privileges, the plan or a plan feature itself. Numbers are plan quotas, not flags; permission on one specific record stays in app code. Full rule: get_integration_guide topic=fit-together. The result carries `warnings` when the flag cannot mean what it looks like — switched on while its onValue is false, or a rule on a plan or role the app does not have. A warning never blocks the write; read it and fix the flag. Before writing app code that consumes this, call get_integration_guide with topic=feature-flags and the framework the app uses.

  ConnectorOAuth

• approve_briefA

  RankwiseA

  Approve a draft brief. This debits the workspace's credits (see get_plan_and_credits for the cost) and starts writing the AI draft article, which then waits for review; nothing is published. Re-approving a draft whose generation failed is free. Send a unique idempotency_key and reuse it on retry.

  Connector

  Destructive

  No auth

• rating_score_writingA

  website-searchA

  Get Lenny Zeltser's scoring playbook so your AI can score a draft locally against a cybersecurity-writing rating sheet. THIS IS THE ONLY TOOL THAT PRODUCES NUMERIC SCORES — the writing-coach tools (`get_security_writing_guidelines`, `ir_*`, `product_*`) never score. Returns the rubric plus step-by-step instructions for applying it. This server never requests your draft and instructs your AI to keep it local—rating sheets and scoring instructions flow to your AI.

  ConnectorNo auth

• malware_get_guidelinesA

  website-searchA

  Get Lenny Zeltser's expert malware analysis report writing guidelines. Topics include capabilities, confidence, pyramid_of_pain, anti_patterns, methodology, fields, handoffs, frameworks, plus tone, words, structure, and executive_summary topics that defer to `get_security_writing_guidelines` for canonical Five Elements guidance. Pair the 'fields' topic with field_id for single-field guidance. This server never requests your sample, analysis notes, or indicators and instructs your AI to keep them local—guidelines and the report template flow to your AI for local analysis.

  ConnectorNo auth

• get_discoveryA

  ikeytz - Schlüsseldienst LudwigsburgA

  WHAT: Fetch one public discovery file from www and return a line window. REQUIRED which. IDs (aliases folded): llms, llms-full, llms-index, llms-keywords, llms-keywords-brand, llms-serp, llms-jsonld, llms-impressum-kontakt, llms-orte-geo, llms-grenzen-koordinaten, llms-urheberrecht, llms-copyright, llms-mcp-server, llms-mcp-web, robots, sitemap-txt, sitemap-xml, ai-txt, ai-plugin, answer-engine, ard, ai-catalog, auth-md, mcp-readme, agent-skills (+ locale slices llms-keywords-en|… ). summary = the line window + trailing # MCP-META (truncated, totalUrls, nextOffset). Omit limit: if remaining >~30KB → first ~30KB window + nextOffset (O9); explicit limit keeps line window 1..10000 (not forced to 30KB). Unknown which → unknown_discovery. Prefer dedicated get_llms_* / get_sitemap_txt when you know the file. Policy files say ai-train=no.

  ConnectorNo auth

• generate-uml-diagramA

  docuwriterA

  Use this when the user supplies source code and asks for UML, Mermaid, sequence, class, component, or architecture diagrams. After authorization, this state-changing generation consumes AI credits, creates a generation, and can save markdown to a Space. Public Spaces expose saved diagrams. Use browser guidance for local uploads or full-repository diagrams.

  ConnectorOAuth

• get_form_logicA

  FaldaroA

  Fetch a form draft’s rule documents — logic (calculations), workflow (states and transitions) and settings — without the bulky field schema. Read this BEFORE writing logic or workflow with update_form_draft, because a document you save replaces that document wholesale. Requires form:write.

  ConnectorAPI key

• get_design_rulesA

  standby-design-mcpA

  Universal design rules for building UI with design tokens: layout, flow (empty, loading and error states, dialogs), shape, surfaces, interaction and keyboard access, forms, content structure, states, typography, color and stack rules for Tailwind/React Native. The tokens say which values exist; these rules say how to use them. Call this before building or reviewing views. Without arguments it returns every rule as one line (62 rules); pass rules (slugs) or a category with detail "full" for the complete text including how to detect violations in code. Rule texts are in German. Categories: Layout, Fluss, Form, Fläche, Bedienung, Formular, Inhalt, Zustand, Typografie, Farbe, Stack.

  ConnectorNo auth