microsoft/cat-agent-skills is a static website that catalogs reusable instruction sets and related packages for AI agents. People use it to search, filter, rate, and download skills for Cowork, Copilot Studio, and Scout, along with Copilot plugins and Scout automations. The catalogue entries are the skills, instructions, plugins, and settings displayed by the site.
Prioritizes the fault by asset criticality and downtime cost into a P1-P4 maintenance priority with a response target and cost estimate, deterministically. Use when the user says "how urgent is this", "what's the priority", "rank by criticality", "what will downtime cost", or after failure-mode-rank completes.
Ranks the likely failure modes for a fault against the failure-mode library, symptoms, alarm codes and repeat-failure history, deterministically. Use when the user says "what's the likely cause", "rank the failure modes", "diagnose this fault", "is this the real root cause?", or after history-retrieve completes.
Reads maintenance fault notes, operator reports, alarm/event logs, asset IDs and location into the mfg.maintenance-triage.v1 contract inputs. Use when the user says "triage this fault", "work this breakdown", "read the fault note for ", "what's wrong with pump ", or when a maintenance triage run begins.
Retrieves the asset's maintenance history — prior work orders, similar failures, PM records — and the relevant OEM manual and troubleshooting sections, and attaches them to the mfg.maintenance-triage.v1 contract. Use when the user says "pull the history for ", "has this failed before?", "find the manual section", "any…
Drafts the work-order update and recommended action from the triaged contract payload, with every determination cited to the manual, history and rules. Use when the user says "draft the work order", "write up the WO update", "what should we do", "recommend the fix", or after criticality-rank completes.
Queues validation exceptions with reason codes and a recommended correction per exception, deterministically. Use on "route the exceptions", "what needs fixing before entry", after order-validate.
Normalises extracted order lines - whitespace, casing, UOM tokens, qty formats - without changing meaning, preparing them for validation. Use after order-ingest.
Ingests orders arriving as email, PDF or portal export into the rtl.order-intake-o2c.v1 contract. Use when the user says "order came in by email", "key in this PO", "process the attached order", or an order document arrives.
Validates the order against customer, pricing, product and credit masters deterministically with the ordervalidate engine. Use on "is this order clean", "validate before entry", after line-normalize.
Drafts the customer confirmation or clarification message and the internal entry note from the routed payload. Use to close every order run - "draft the reply to the customer", "write up the order status".
Normalises supplier fields to the taxonomy and validates GTIN check digits with the deterministic attributenormalize engine. Use after source-ingest, or on "map these to our taxonomy", "are these GTINs valid".
Validates marketing claims and regulated terms against the approved claim library - the explicit Govern step - with the deterministic claimcheck engine. Use on "is this copy safe", "can we say antibacterial", "check the claims", after gap-score.
Drafts channel descriptions, comparison copy and the review/approval packet from the governed payload. Use to close every enrichment batch: "draft the product copy", "build the review packet".
Scores per-SKU completeness against the channel's required fields with the deterministic completenessscore engine. Use on "how complete are these", "what's missing per SKU", after attribute-normalize.
Ingests supplier spreadsheets, spec sheets and copy into the rtl.product-content-enrichment.v1 contract. Use when the user says "enrich these SKUs", "supplier sent the item sheet", "get these products channel-ready", or an enrichment batch begins.
Checks capacity, constraints and material availability for the plan week with the deterministic capacitycheck engine. Use when the user says "do we have capacity", "does the week fit", "check materials", or after demand-pull in a planning run.
Pulls the demand forecast and open orders for the plan week into the mfg.production-planning.v1 contract. Use when the user says "plan next week", "build the schedule", "pull the orders for line ", or when a planning run begins.
Optimizes the weekly sequence for changeovers with the deterministic campaign heuristic - naive vs optimized compared side by side - using the sequenceoptimize engine. Use when the user says "optimize the schedule", "sequence the week", "minimize changeovers", or after capacity-check in a planning run.
Drafts the shift plan from the optimized schedule for the planner to review and publish. Use when the user says "draft the shift plan", "write up the week", "publish the plan" (which produces the draft), or after schedule-optimize in a planning run.
Builds corrective and preventive actions with mandatory effectiveness checks from the verified root cause, using the deterministic capalogic engine. Use when the user says "build the CAPA", "what actions do we take", "corrective actions please", or after root-cause-analyze in a CAPA run.
Validates CAPA closure readiness against the closure rules - the explicit Govern step of this plugin. Use when the user says "can we close this CAPA", "is the 8D ready to close", "close it out", or before any closure recommendation.
Drafts the 8D / CAPA report from the contract payload, cited to the clause and rules, with open blockers leading. Use when the user says "draft the 8D", "write the CAPA report", or after closure-check in a CAPA run.
Reads the open NCR (the mfg.quality-inspection.v1 payload handed off by the Quality Inspection plugin), complaint log and inspection data into the mfg.quality-incident-capa.v1 contract. Use when the user says "open a CAPA for NCR ", "work the nonconformance", "start the 8D", or when a CAPA run begins.
Establishes root cause with a structured method - Pareto over the complaint window, then an evidenced 5-Why chain - using the deterministic pareto and rcatree engines. Use when the user says "what's the root cause", "run the 5-why", "why does this keep happening", or after ncr-intake in a CAPA run.
★not rated 93▲
+4 todayA82 tokens
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: