microsoft/cat-agent-skills

Skills for modern agents in Copilot Studio

About the project

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.

This repository also configures its own agents. See what cat-agent-skills tells them →

93Stars on the repository
194Mods indexed here, across every type
todayLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

pack-ingest

169

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Ingests the HQ campaign pack, price file, planogram and task list into the rtl.store-task-promotion-execution.v1 contract. Use when the user says "new campaign pack landed", "prep the store for the launch", "are we ready for Monday's promo", or when a store assignment arrives.

not rated 93 +4 today A 68 tokens original MIT

readiness-check

170

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Tests store readiness against the campaign pack deterministically - applicability, completeness, fixture conflicts and price integrity - with the readinesscheck engine. Use when the user asks "are we ready", "any gaps for launch", "check the pack against my store".

not rated 93 +4 today A 54 tokens original MIT

store-map

171

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Retrieves the store profile - format, cluster, fixtures, received assets, completed tasks, prior execution history - for the readiness test. Use after pack-ingest, or when the user asks "what does this store actually have".

not rated 93 +4 today A 49 tokens original MIT

disruption-intake

172

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Picks up a supply disruption from a PO record, shipment feed, supplier email or portal notice and pulls it into the mfg.supplier-disruption.v1 contract inputs. Use when the user says "a PO is late", "our supplier is delayed", "we're short on a part", "process this supplier notice", "shipment is going to slip", or when…

not rated 93 +4 today A 88 tokens original MIT

disruption-writeup

173

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Drafts the supplier follow-up and the internal escalation brief from the assembled contract payload, with the impact and mitigation cited to the ERP/MRP, supplier master and rules. Use when the user says "draft the supplier email", "write the escalation brief", "draft the follow-up", "give me the escalation", "write…

not rated 93 +4 today A 80 tokens original MIT

mitigation-recommend

174

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Chooses the mitigation (expedite, substitute or reschedule) for a mapped shortage and computes the escalation tier and owner-notification list, deterministically. Use when the user says "what do we do about it", "should we expedite or substitute", "recommend a mitigation", "who do we escalate to", "who do we notify"…

not rated 93 +4 today A 82 tokens original MIT

shortage-impact-map

175

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Maps a part shortage through BOM/where-used to the affected products, production lines and open orders, with the quantity and date at risk per order, deterministically. Use when the user says "what does this shortage affect", "which lines and orders are hit", "when does this bite", "map the impact", "what's at risk"…

not rated 93 +4 today A 84 tokens original MIT

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Retrieves the ERP/MRP supply position, supplier master, BOM/where-used and open orders relevant to a disruption and attaches them plus the supplier's delivery history to the mfg.supplier-disruption.v1 contract. Use when the user says "what does this part go into", "where is this used", "what's our on-hand and lead…

not rated 93 +4 today A 101 tokens original MIT

avl-compare

177

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Compares the candidate against the approved vendor list and category requirements and produces the banded recommendation, using the deterministic avlmatch engine. Use when the user says "how do they compare to our current vendors", "check the AVL", "should we qualify them", or after risk-score in a run.

not rated 93 +4 today A 66 tokens original MIT

dossier-ingest

178

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Gathers the supplier's qualification documents - certificates, financials, audit reports, PPAP elements - into the mfg.supplier-qualification.v1 contract. Use when the user says "qualify this supplier", "review the supplier package", "new vendor for ", or when a qualification run begins.

not rated 93 +4 today A 68 tokens original MIT

qualification-memo

179

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Drafts the supplier qualification memo for approval from the contract payload - outcome, conditions, risk detail, AVL position. Use when the user says "draft the qualification memo", "write it up for approval", or after avl-compare in a run.

not rated 93 +4 today A 55 tokens original MIT

risk-score

180

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Scores supplier risk across quality, financial and geographic dimensions with the deterministic riskscore engine. Use when the user says "score this supplier", "how risky are they", "run the risk assessment", or after dossier-ingest in a qualification run.

not rated 93 +4 today A 52 tokens original MIT

issue-cluster

181

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Clusters recurring issues by type and lane with record evidence, and maps scorecard breaches to governing trading terms, deterministically. Use on "what keeps going wrong", "cluster the issues", after scorecard-roll.

not rated 93 +4 today A 46 tokens original MIT

record-pull

182

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Pulls PO, receipt, invoice, return, quality and dispute records for the vendor and review period into the rtl.supplier-vendor-performance.v1 contract. Use when the user says "prep the vendor review for ", "pull the supplier records", "quarterly business review coming up".

not rated 93 +4 today A 64 tokens original MIT

review-pack

183

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Drafts the vendor review agenda, corrective action request and commitment tracker from the contract payload. Use to close every review prep: "draft the review pack", "build the QBR agenda".

not rated 93 +4 today A 41 tokens original MIT

scorecard-roll

184

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Rolls up the OTIF / defect / dispute scorecard by quarter with trend flags, deterministically, with the scorecardroll engine. Use on "how are they performing", "roll the scorecard", after record-pull.

not rated 93 +4 today A 51 tokens original MIT

term-retrieve

185

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Retrieves and quotes the governing contract or trading term for any performance question. Use on "what does the agreement say about OTIF", "do we have a rebate right", or after issue-cluster.

not rated 93 +4 today A 45 tokens original MIT

claim-ingest

186

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Extracts deduction records and claim backup (PDFs, portal exports) into the rtl.trade-deduction-recovery.v1 contract. Use when the user says "retailer deducted from the invoice", "work this deduction", "claim backup arrived", or a deduction case begins.

not rated 93 +4 today A 60 tokens original MIT

claim-match

187

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Matches each claim to its governing promotion or trade term deterministically with the claimmatch engine. Use on "does this deduction match anything we signed", after term-retrieve.

not rated 93 +4 today A 37 tokens original MIT

dispute-packet

188

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Drafts the dispute packet and the post-event promotion brief from the governed payload. Use to close every deduction case: "draft the dispute", "build the packet for the retailer".

not rated 93 +4 today A 41 tokens original MIT

validity-classify

189

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Classifies deduction validity and quantifies the disputable amount with the calculation shown - the explicit Govern step - and runs promotion lift only where the RGM scoping allows. Use on "is this deduction valid", "how much can we dispute", "should we write it off", after claim-match.

not rated 93 +4 today A 65 tokens original MIT

parts-readiness

190

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Checks whether a work order's parts are ready — storeroom stock, shortages, and part supersession — and computes a READY/PARTIAL/BLOCKED status, deterministically. Use when the user says "are the parts in stock", "check parts readiness", "can we schedule this", "is that part still current", or after similar-work-rank…

not rated 93 +4 today A 79 tokens original MIT

similar-work-rank

191

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Ranks the most similar past work orders for the current job and surfaces recurring failures and superseding fixes, deterministically. Use when the user says "find similar past work", "have we done this before", "what's the closest prior fix", "is this a repeat failure", or after knowledge-retrieve completes.

not rated 93 +4 today A 68 tokens original MIT

wo-intake

192

microsoft/cat-agent-skills

Skill Claude CodeCodex ✓ vendor

Pulls the current work order and its asset context into the mfg.work-order-assist.v1 contract inputs. Use when the user says "assemble the work order packet", "pull up WO ", "get me everything for this work order", "work order for ", or when a Work Order Assistant run begins.

not rated 93 +4 today A 71 tokens original MIT

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: