prism-shadow

@prism-shadow source repo

997 skills
0 followers
13 hours ago last updated

▌
Reflect 3 Attempt 03 62 · prism-shadow
Atlas Commerce Operations — Analytical Task Skill

0
installs
▌
Reflect 3 Attempt 01 63 · prism-shadow
Public Health Observatory Audit Skill

0
installs
▌
Reflect 3 Attempt 02 65 · prism-shadow bundle
Public Health Observatory Algorithmic Audit Skill

0
installs
▌
Reflect 3 Attempt 03 63 · prism-shadow bundle
Public Health Observatory Algorithmic Audit Skill

0
installs
▌
Reflect 3 Attempt 01 64 · prism-shadow
Portfolio Engineering Skill

0
installs
▌
Reflect 3 Attempt 02 66 · prism-shadow
Portfolio Engineering Skill

0
installs
▌
Reflect 3 Attempt 03 64 · prism-shadow
Portfolio Engineering Analysis Skill

0
installs
▌
Harborcrm Front Of Funnel 3 · prism-shadow
Solve HarborCRM front-of-funnel CRM/marketing tasks against the read-only HarborCRM JSON API: post-event sponsor reconciliation and CRM handoff, trade-show exhibitor prospecting/qualification, raw contact-import hygiene (dedupe/suppress/normalize), sponsor/lead follow-up planning, and CRM create/update/no-action decisions. Use whenever a task references HarborCRM, an event_id (sponsor orders/badges/invoices), a trade-show show_id (exhibitors/meeting_interest), or an import batch of raw contacts, and asks for a structured JSON answer matching an answer_template.

0
installs
▌
Harborcrm Front Of Funnel 4 · prism-shadow
Solve HarborCRM front-of-funnel CRM/marketing tasks against the read-only HarborCRM JSON API: post-event sponsor reconciliation, finance/sponsor-status classification, trade-show prospecting & lead qualification, raw contact-import hygiene/dedup/suppression, badge reconciliation, follow-up due-date and task planning, and CRM create/update/no-action decisions. Use whenever a task references HarborCRM, an event_id, a trade-show show_id, an import batch, sponsor packages/invoices, badge scans, exhibitors, meeting interest, or asks for a JSON handoff/prospecting/import summary matching an answer_template.

0
installs
▌
Harborcrm Front Of Funnel 5 · prism-shadow
Solve HarborCRM front-of-funnel CRM/marketing tasks against the shared read-only HarborCRM JSON API. Use this whenever a task references HarborCRM, an event_id / show_id / import batch_id, sponsor reconciliation, badge scans, trade-show exhibitor prospecting/qualification, raw contact-import hygiene/dedup/suppression, sponsor or lead follow-up planning, or CRM create/update/no-action decisions — even if the task only says "reconcile the event", "prepare the qualified lead list", "clean the import batch", or "prospecting summary". The output is always a single strict JSON object matching the provided answer_template; this skill explains the business rules (source precedence, sponsor status enums, normalization, exclusion logic, follow-up due dates, rollups) that determine the correct values.

0
installs
▌
Medbridge Sales Ops 2 · prism-shadow bundle
Produce account-ready JSON answers for the MedBridge Sales Ops domain (CRM, quotes, freight/logistics decisions, and milestone receivables). Use this whenever a task asks you to verify or reconcile MedBridge records against the Sales Ops HTTP API and return JSON matching an answer_template.json — including catalog-tier quote pricing, EXW + freight decision packages, route/border risk and recommended transport mode, module-only RFQ quotes, payment-terms selection, or opportunity↔invoice↔payment↔revenue-journal reconciliation with collection/event follow-up routing. Trigger it even when the prompt only names a quote/opportunity/RFQ id (e.g. "Q-...", "OPP-...", "RFQ-...") and asks for a decision package or reconciliation, or mentions freight options, milestones, outstanding balance, revenue recognition, vouchers, or follow-up tasks.

0
installs
▌
Medbridge Sales Ops 3 · prism-shadow
Produce account-ready JSON decision packages from the MedBridge Sales Ops API for CRM / quote / logistics / receivables tasks. Use this whenever a task asks you to build or verify a quote decision package, freight/transport comparison, EXW or module-level quote, catalog-tier pricing, route/border risk and freight validity flags, recommended transport mode, payment terms, or an opportunity/invoice/payment/revenue-recognition reconciliation with CRM follow-up routing — especially when the prompt references a MedBridge / "Sales Ops" API base URL (e.g. API_BASE_URL / BASE_URL / http://127.0.0.1:<PORT>), quote IDs (Q-...), RFQ IDs (RFQ-...), freight IDs (FR-...), opportunity IDs (OPP-...), invoices (INV-...), or asks to "return only JSON matching input/payloads/answer_template.json". Trigger even if the user only describes the business outcome (e.g. "reconcile this account", "compare freight options", "indicative module quote") without naming the API.

0
installs
▌
Crm Support Ticket Resolution 2 · prism-shadow bundle
Resolve CRM service-ticket / support-queue tasks against the shared read-only support-console API. Use this whenever a task asks you to act as a support operations analyst, contact-center lead, queue-quality analyst, mobile-data recovery analyst, or enterprise support lead and to "resolve", "classify", "route", "triage", or "produce a structured response/summary" for tickets, cases, mobile lines, or enterprise export incidents, returning JSON that conforms to a provided answer_template.json. Trigger it for any payload of service tickets, a case/queue snapshot (CSV or JSON), a mobile worklist, or an enterprise complaint email, even if the prompt does not name this skill — the decisions must come from support-console records, never from assumptions.

0
installs
▌
Crm Support Ticket Resolution 3 · prism-shadow bundle
Resolve CRM support-console tasks by reading the shared read-only support console API and emitting the exact JSON the answer_template demands. Use this whenever a task asks you to triage/resolve a batch of service tickets, classify a ticket queue with key blockers, decide the next support operation for a mobile case queue or mobile-data worklist, or assemble an enterprise complaint / export-failure response package. Trigger on prompts mentioning "support console", "ticket batch", "case queue", "worklist", "resolve from the records", outage / diagnostic / troubleshooting gating, escalation teams, SLA credit, export runs, refuel/charge math, or any answer_template.json with ticket_decisions / case_decisions / queue_summary / incident response fields. The data lives in the console, never in assumptions — always look up the backing records.

0
installs
▌
Apexcloud Retention Ops 6 · prism-shadow bundle
Standard operating procedure for the ApexCloud Retention Operations API (http://127.0.0.1:8074). Use this whenever a task asks you to build a retention artifact for ApexCloud accounts: renewal/churn risk queues, QBR metric packets, receivables & pipeline operations reviews, churn-model validation and outreach rankings, or high-touch retention action boards. Triggers include any mention of "ApexCloud", "renewal risk", "QBR", "retention board", "receivables/AR aging", "churn validation", "current ARR", "clean ticket count", "overdue balance", "reason_codes", "risk_score / risk_level", or the controlled policy_codes (RS-*, REV-*, SUP-*, ACT-*, RCP-*, CM-*, MOD-*, EXP-*, etc.). Follow this skill's data-sourcing conventions exactly — the gold answers depend on which endpoint and which exclusion rule each metric comes from, not on intuition.

0
installs
▌
Apexcloud Retention Ops 7 · prism-shadow bundle
Operating procedure for the ApexCloud Retention Operations API (http://127.0.0.1:8074). Use this skill WHENEVER a task asks you to build, score, or summarize anything in the ApexCloud retention / customer-success / revenue-ops domain: renewal risk queues, retention action boards, QBR metric packets, receivables + pipeline operations reviews, churn-model validation, outreach rankings, or board/leadership summaries. Trigger it even when the request only mentions pieces of this world (current ARR, "clean" ticket counts, latest NPS, overdue/A-R aging buckets, risk_score / risk_level, primary_action, reason_codes, expansion pipeline, win rate, policy_codes like RS-6 / REV-4 / SUP-8 / ACT-5 / RCP-7 / CM-5 / MOD-7, or account ids like acct_*). These tasks share fixed business conventions that are easy to get wrong from intuition; this skill encodes the exact, verified rules.

0
installs
▌
Apexcloud Retention Ops 8 · prism-shadow bundle
Operating procedure for solving ApexCloud Retention Operations tasks against the read-only HTTP API at http://127.0.0.1:8074. Use this skill whenever a task asks you to build a renewal risk queue, a retention action board, a QBR metrics packet, a receivables / pipeline operations review, or a churn-model validation & outreach ranking for ApexCloud accounts — or any time the prompt mentions current ARR, billing snapshots, clean ticket counts, latest NPS, overdue A/R aging buckets, risk scoring, reason codes, primary actions, or the four-letter policy_codes (RS-/REV-/SUP-/ACT-/RCP-/CM-/PW-/FS-/MOD-/PRB-/DEP-/OUT-/BORD-/EXP-/CAL-). These tasks share fixed business conventions that are easy to get wrong; follow this skill instead of guessing the data sources, exclusion rules, thresholds, or enums.

0
installs
▌
Procureops Erp Answers 2 · prism-shadow bundle
Produce the required JSON answer for a ProcureOps ERP "Procurement Supplier & Receiving Control" task — sourcing nomination readiness, receiving/AP reconciliation, AP close hold-or-release with vendor balances, change-control contract & budget-headroom decisions, and AP release with chargeback netting. Use this whenever a task hands you a ProcureOps API base URL plus a local memo/payload and an answer_template.json and asks for a decision packet, close file, reconciliation, readiness review, release/hold queue, nomination decision, or vendor-risk-as-of analysis over programs, suppliers, items, contracts, requisitions, purchase orders, receipts, AP invoices, payments, approvals, budget snapshots, or vendor-risk events. Trigger even when the word "ProcureOps" is absent but the inputs are an ERP procurement/AP/receiving API + a memo + a JSON template to fill.

0
installs
▌
Northwind Erp Ops 2 · prism-shadow bundle
Solve Northwind Components ERP operations tasks by querying the live read-only ERP API and returning the exact JSON decision file the task asks for. Use this whenever a task references the Northwind ERP, a wave/expedite/allocation/transfer/replenishment/kit-build queue, a supplier incident or quality scorecard, a procurement/replenishment-control review, a shipping quote, or any prompt that hands you a memo/request payload plus an answer_template.json and asks for fulfillment, allocation, backorder, transfer, purchase-requisition, hold/freeze, or recommendation decisions. Trigger even if the task only mentions SKUs (NW-####), orders (SO-#####), suppliers (SUP-###), POs (PO-#####), incidents (INC-#####), warehouses (WH_NORTH/WH_CENTRAL/WH_WEST), BOMs, or "effective available" / "safety stock" inventory math.

0
installs
▌
Northwind Erp Ops 3 · prism-shadow bundle
Solve Northwind Components ERP operations tasks against the shared read-only ERP API: expedite/dispatch decisions, mixed-warehouse allocation and transfer waves, kit-build replenishment (BOM coverage, transfers, purchase requisitions), supplier incident scorecards, and supplier quality-hold / replenishment-control reviews. Use this skill whenever a task references the Northwind Components ERP, a "wave", an expedite/allocation/ dispatch queue, a BOM/kit build, a supplier incident scorecard, a quality-hold review, effective inventory availability, shipping quotes, or asks for a structured JSON decision file built from /orders, /products, /inventory, /customers, /suppliers, /warehouses, /boms, /purchase_orders, /incidents, or /shipping/quote — even if the task only mentions a memo, planning extract, or answer_template.json. The local memo/extract is planning context only; the live API is the source of truth.

0
installs
▌
Private Wealth Advisory Planning 3 · prism-shadow bundle
Produce structured JSON planning outputs for a private-wealth advisory benchmark backed by a read-only advisory API (Roth-conversion/RMD tax summaries, ILIT/Crummey funding cycles, GRAT-vs-CRAT trust comparisons, and integrated estate-liquidity action plans). Use this skill whenever a task asks for a structured advisory planning JSON for a client ID (e.g. CLT-xxxx), references an answer_template.json with fields like conversion_plan, rmd_projection, gift_plan, administration, grat/crat, estate_context, ilit, trust_transfer, action_set, or source_resolution, or mentions Roth conversions, RMDs, ILIT Crummey notices, GRAT, CRAT, estate-tax exemption, or estate liquidity. It encodes the exact source-conflict resolution rules, numeric formulas, RMD year-by-year simulation order, date arithmetic, enum decision rules, rounding, and output schema the gold answers require, so apply it even when the prompt does not name a specific strategy by name.

0
installs
▌
Private Wealth Advisory Planner 2 · prism-shadow bundle
Produce the structured JSON planning output for the Private Wealth Advisory benchmark — Roth-conversion/RMD tax summaries, ILIT/Crummey funding cycles, GRAT-vs-CRAT trust comparisons, and integrated estate-liquidity action plans. Use this whenever a task references a private-wealth / estate / advisory client (e.g. "CLT-####"), an advisory request memo, an answer_template.json that must be filled, a read-only advisory API (clients, source-documents, retirement-accounts, life-insurance, trust-candidates, policies/tax, rmd-factors), or asks for Roth conversion sizing, RMD projections, Crummey notice timing, gift-tax exclusion capacity, estate-tax exposure, GRAT/CRAT remainders, or an estate action set. Trigger even when the words "skill" or the analysis type are not stated explicitly.

0
installs
▌
Crescent Finance Ops Reporting 2 · prism-shadow bundle
Use this skill for any Crescent Finance Ops / Crescent Arts Collective management-reporting task that returns a single JSON object built from the read-only Finance Ops HTTP API. It covers branch monthly-close packages, regional rollups, current-year and forecast compensation summaries, and weekly theatre payroll / CBA control. Trigger it whenever the prompt mentions a request_memo + answer_template + environment_access payload, a Finance Ops base_url, income statement / EBITDA / ARPU / sales-per-labor-headcount, branch or region rankings, compensation by quarter and pay type, seniority or overscale, or weekly payroll with services, doubles, premiums, guarantees and conflict flags. Even when the task only says "prepare the close package" or "review this payroll week", use this skill — the numeric conventions and business rules here are where solvers most often go wrong.

0
installs
▌
Crescent Finance Ops Reporting 3 · prism-shadow bundle
Produce Crescent Finance Ops management-reporting deliverables for Crescent Arts Collective: branch monthly-close packages, regional reporting views, current-year and forecast ensemble compensation summaries, and theatre weekly payroll / CBA control reviews. Use this whenever a task references the Crescent Finance Ops API (finance branches/records, compensation rate-book/rosters/ scenarios, payroll rate-book/productions), a request_memo + answer_template pair, branch/region income statements, EBITDA / ARPU / sales-per-headcount ratios, MoM or FY growth, musician/ensemble pay by quarter and pay type, weekly payroll category totals, or CBA conflict flags — even if the word "Crescent" is not used but the payloads or endpoints match. Reach for it before computing by hand: the rounding and scoping conventions here are easy to get wrong.

0
installs
▌
Crescent Finance Ops Reporting 4 · prism-shadow bundle
Produce Crescent Finance Ops management-reporting deliverables (Crescent Arts Collective) by querying the read-only Finance Ops HTTP API and returning one JSON object that matches a provided answer_template. Use this skill WHENEVER a task involves Crescent / Finance Ops / Crescent Arts Collective and any of: a branch monthly-close package or income statement; MoM revenue variance; FY2024-vs-FY2025 branch or regional comparisons; EBITDA / ARPU / sales-per-labor-headcount ratios; branch or regional EBITDA rankings; ensemble compensation by quarter and pay type; multi-year board compensation forecasts with scenario escalation; or theatre / touring-production weekly payroll, musician pay categories, doubles/overscale/ substitute/guarantee handling, or CBA conflict flags. Trigger even if the user only mentions a request_memo, answer_template, an ENS-/BR-/REG-/PROD- id, or a "close package / management package / board forecast / payroll review", and does not name the API explicitly.

0
installs
▌
Asteria Investment Office JSON 5 · prism-shadow
Produce strict-JSON deliverables for the Asteria Investment Office (CIO / credit desk) task family: energy/fixed-income credit trade packages, international equity correlation reviews, cross-asset active-allocation view refreshes, fixed-income risk-rebalance rotations, and combined committee decision files. Use this whenever a prompt references an Asteria portfolio (PF-*), the shared Asteria HTTP/JSON environment, an answer_template.json output contract, and asks for trade packages, correlation/concentration findings, allocation views (UW/N/OW), or rotation/rebalance JSON. Covers which API endpoints to call, exact formulas/rounding, controlled-enum mapping, and current-environment-over-stale-payload precedence.

0
installs
▌
Credit Committee Packets 2 · prism-shadow bundle
Produce committee-ready JSON answers for the shared Credit Office / lending-committee task family backed by the HTTP API at http://127.0.0.1:8003. Use this skill WHENEVER a prompt asks you to prepare a credit-risk, lending-committee, or credit-union-segment packet against that environment — including branch loan-rating migration / regrade reviews, watch-list stress and workout queues, pending-application capacity & concentration allocations, competing CRE underwriting decisions, and credit-union segment posture pages. Trigger on cues like "rating migration", "regrade", "watch-list stress", "+200bp DSCR stress", "lending-committee allocation", "concentration flags", "decline reason codes", "weighted CRE / CDFI score", "segment posture", "branch_id", "segment_id", "answer_template.json", or any request that says to follow input/payloads/answer_template.json and return only JSON. Always read the live /api/policies endpoint for authoritative thresholds rather than guessing numbers.

0
installs
▌
Credit Committee Packets 3 · prism-shadow bundle
Produce committee-ready JSON answers for the shared Credit Office / lending-committee task family: branch loan-rating regrades and migration reviews, branch lending-capacity allocation packages for pending applications, credit-union segment posture pages, watch-list stress and workout packets, and competing CRE underwriting decisions. Use this skill whenever a prompt references a Credit Risk Committee, lending committee, branch_id or segment_id, risk-rating regrade/migration, watch-list/adverse-rated loans, CDFI risk classes, DSCR stress (+200bp / dual stress), NPA or delinquency benchmark variance (FDIC/NCUA), sector/CRE concentration limits, lending capacity allocation, or asks for a JSON answer matching an answer_template.json against the credit office HTTP API at http://127.0.0.1:8003. Apply it even when the prompt only hints at these (e.g. "regrade the loans rated 3 or worse", "compare the two CRE requests", "posture recommendation for the segment") and does not name the skill explicitly.

0
installs
▌
Peopleops Console Reconciliation 2 · prism-shadow bundle
Solve PeopleOps Console HR-lifecycle reconciliation tasks against the local JSON API (employees, cases, payroll/leave ledgers, recruitment, documents, messages/notices, audit events). Use this whenever a task asks you to verify an onboarding closeout, validate leave or payroll source precedence, review a case folder and formal notice, reconcile a recruitment outcome packet, or pick the controlling audit event for an employee/case/opening — i.e. any prompt that says "use the authoritative submitted/approved record not the draft", "fill in the answer template", references EMP-/CASE-/REQ-/PAY-/LA-/OFFER-/AUD- IDs, or wants normalized business labels rather than free text. Reach for it even when the prompt only describes the scenario without naming the console explicitly.

0
installs
▌
Peopleops Console Review 2 · prism-shadow bundle
Use this skill whenever a task asks you to review, verify, reconcile, or validate something in an HR "PeopleOps Console" / People Ops lifecycle portal exposed as a JSON API — for example onboarding closeout, leave source precedence, payroll/accrual readiness, policy-case folder and formal-notice review, recruitment outcome reconciliation, or audit-scope selection. Trigger it any time the prompt names a PeopleOps Console, an employee/case/opening/ payroll-ledger/leave-assignment/audit record, and asks for a JSON answer that must use normalized business labels from an answer_template.json. It encodes the standard operating procedure: which endpoints to query, the source-precedence and gating rules, the notice-defect vocabulary, and how to produce output that exactly matches the template's enum contract.

0
installs
▌
Self Attempt 02 2 · prism-shadow
MedBridge Sales Ops SOP

0
installs
▌
Crm Service Ticket Sop 2 · prism-shadow
Resolve CRM support-console service tickets, mobile cases, fixed-service queue quality reviews, and enterprise export complaints into strict JSON answer templates using staged payloads plus support-console evidence.

0
installs
▌
Apexcloud Retention Ops 2 · prism-shadow
Use this skill for ApexCloud Retention Operations API tasks that ask for renewal risk queues, QBR metric packets, receivables and pipeline reviews, churn validation/ranking, or high-touch retention action boards. It gives the endpoint habits, filtering rules, source priorities, controlled enum conventions, precision rules, and common pitfalls needed to produce JSON-only answers from the public ApexCloud environment.

0
installs
▌
Apexcloud Retention Ops 3 · prism-shadow
Use this skill for ApexCloud Retention Operations API tasks that ask for customer success, renewal risk, QBR metrics, receivables, pipeline, churn-model validation, or retention action-board JSON outputs. It is especially useful when prompts mention ApexCloud, account metrics, support tickets, NPS, billing snapshots, A/R aging, opportunities, HR/event context, churn exports, controlled labels, policy codes, or answer templates; use it even when the user only asks for the final JSON.

0
installs
▌
Self Attempt 02 3 · prism-shadow
ERP Finance Control Review

0
installs
▌
Procureops Reconciliation 2 · prism-shadow
Solve ProcureOps procurement, receiving, AP, contract-change, and payment-release reconciliation tasks using the live API plus task-local payloads, with exact JSON outputs.

0
installs
▌
Northwind Erp Control Desk 2 · prism-shadow
Use this skill for Northwind Components ERP evaluation tasks that ask for strict JSON decisions using the shared ERP API, especially dispatch expedite queues, mixed-warehouse allocation, kit replenishment, supplier incident scorecards, procurement quality holds, inventory coverage, purchase orders, BOMs, shipping quotes, customers, suppliers, and products. It gives SOPs for API usage, field conventions, stock math, decision precedence, ordering, rounding, and common pitfalls.

0
installs
▌
Agenthub Python · prism-shadow bundle
Guidance for using the AgentHub Python SDK (`agenthub-python`). Use when developing agents that call different LLM APIs, need a unified interface for LLM providers, mention AgentHub, request `agenthub-python`, or already import it.

0
installs
▌
Agenthub Typescript · prism-shadow bundle
Guidance for using the AgentHub TypeScript SDK (`@prismshadow/agenthub`). Use when developing agents that call different LLM APIs, need a unified interface for LLM providers, mention AgentHub, request `@prismshadow/agenthub`, or already import it.

0
installs
▌
Agenthub Dev · prism-shadow
Fixed workflow for developing AgentHub itself — adding or updating model support. Use when asked to support a new model or protocol version in this repository, sync llmsdk_docs, or implement a provider client. Covers doc syncing, live API capture, paired Python/TypeScript implementation, and model-scoped e2e testing.

0
installs
▌
Penguin Harness Dev · prism-shadow bundle
Use when developing PenguinHarness itself — changing packages/{core,server,web,cli,desktop,landing,docs,skills}, the built-in model catalog, the installers or the release workflow; writing or auditing changelog entries; writing a blog post or capturing release screenshots; running the test suite here or, when asked, on another machine; deciding what to do about data already on disk; or auditing prose that reads like a leaked authoring session. Covers the two-repo symlink layout, the CI-parity verification chain, the record-and-ship contract, where blog media is hosted, and the seams that are intentional.

0
installs
▌
Humanizer · prism-shadow bundle
Rewrite or edit prose in any language so it reads like edited human writing in the register of books, newspapers and encyclopedias rather than default AI output. A small drafting core — vary every pattern, build density from anchored facts in whole grammar, cap the quotables, put a real writer with real material behind the text, let structure serve content, write each language from inside its idiom and typography, verify what you assert, and aim for the natural distribution of edited prose rather than a perfect scorecard — backed by a three-layer tell catalog, per-language cue files for six languages and a seven-round measured case study shipped as reference files for the diagnostic census.

0
installs
▌
Vllm · prism-shadow
Deploy and serve LLMs with vLLM behind an OpenAI-compatible endpoint, with tool calling enabled for agent workloads.

0
installs
▌
Firecrawl · prism-shadow
Search the web and scrape pages into clean markdown with the Firecrawl API — query-based discovery, single-URL extraction including public PDFs, driven by curl with a vault-stored API key.

0
installs
▌
Penguin Harness Frontend · prism-shadow
Use when changing the PenguinHarness Web App (`packages/web`) — adding or restyling any UI, picking a status colour, adding an icon, laying out a row or a form field, writing user-facing copy, or building a popup. Covers the semantic tone tokens, the icon size/stroke/gap scale, the semantic-versus-formatting rule for explanatory text, the two-dictionary i18n contract, and the portal-panel pattern with its Esc and scroll caveats.

0
installs
▌
Ollama · prism-shadow
Deploy and serve local models with Ollama — pull and run them, then expose the OpenAI-compatible endpoint to apps and agents.

0
installs
▌
Penguin Harness Manual Test · prism-shadow
Use when standing PenguinHarness up to try a change by hand — launching the Web App, the desktop shell, the landing page or the docs site to click through it, screenshot it, or reproduce a report. Covers the four dev entry points and their ports, which data root each writes to, and the four ways a healthy setup looks broken.

0
installs
▌
Data Analysis · prism-shadow
Complete data-analysis tasks with bounded inspection, correct data semantics, native artifact handling, complete delivery, and risk-based verification.

0
installs
▌
Skill Porting · prism-shadow
Install skills from external ecosystems into this agent's agent_state/skills/ — resolve Claude Code plugin marketplaces, the Codex plugin repo, skills.sh registry names, GitHub repos, or local folders to their skill directories, review every file, and normalize SKILL.md frontmatter to the Penguin format.

0
installs
▌
Penguin Sdk · prism-shadow
Use whenever the user wants to build an agent application — their own program with an embedded agent, such as an AI app, an agentic app or a RAG app. This is writing application code on the Penguin Harness SDK, not configuring an Agent State inside PenguinHarness. Covers self-contained projects, the createSession/run streaming loop with thinking and image messages, wiring the user's existing tools in as CLI commands, and a complete RAG recipe that ingests documents into a knowledge base and answers with citations behind a web UI.

0
installs
▌
Agent Evaluation · prism-shadow
Run one specified Test Agent on one specified Benchmark Case exactly once, privately score that execution, and return one protocol result.

0
installs
▌
Benchmark Design · prism-shadow
Design and calibrate a multi-Case capability Benchmark and establish a traceable Formal Baseline.

0
installs
▌
Bento Slides · prism-shadow
Authoring Bento decks

0
installs
▌
Llamafactory · prism-shadow
Fine-tune LLMs with LlamaFactory — register datasets, train via YAML configs, merge LoRA adapters and serve the result.

0
installs
▌
Agent Optimization · prism-shadow
Improve an Agent State through versioned scores and score-linked Traces from a frozen Benchmark.

0
installs
▌
Web Design · prism-shadow
Penguin visual language for generated web pages and app UIs — GitHub-style simplicity with a single blue accent, light and pure-black dark themes, design tokens, component and chat-interface recipes, plus an opt-in warm paper editorial theme.

0
installs
▌
Penguin Config · prism-shadow
Manage model API keys, default models and per-agent vault secrets with the penguin CLI.

0
installs
▌
Unified LLM API · prism-shadow
Call model APIs through @prismshadow/agenthub — streaming text generation, image generation, speech synthesis, embeddings and the supported-model registry with one client.

0
installs
▌
Agent Initialization · prism-shadow
Initialize an Agent's settings from a user requirement by writing AGENTS.md, setting identity metadata, and installing only needed Skills.

0
installs
▌
Remote Claude Code · prism-shadow bundle
Run Claude Code on a remote host over SSH — a persistent expect-driven login session, headless claude -p with the stdin fix, the interactive TUI inside a remote tmux driven by send-keys/capture-pane (one keystroke at a time, capture-verified; relayed user messages go through verbatim), and multi-turn continuity via --session-id/--resume or stream-json; hosts and credentials are placeholders resolved at runtime from the user or the vault, never hardcoded.

0
installs
▌
Penguin Orchestration · prism-shadow
Drive PenguinHarness itself from a shell — list and create agents and sessions, send and steer messages mid-flight, and query costs and scheduled tasks via the penguin CLI over the local server.

0
installs
▌
Software Engineering · prism-shadow
Complete software-engineering tasks — investigate and review code, implement bug fixes, features and refactors with minimal scope, validate changes, and report verified outcomes.

0
installs
▌
Skill Creator · prism-shadow
Create a reusable skill from staged training examples and matching answers.

0
installs
▌
Harborcrm Handoff Solver · prism-shadow
Solve HarborCRM public-API tasks for CRM event handoff, sponsor reconciliation, trade-show prospecting, and import-batch cleanup.

0
installs
▌
Self Attempt 02 · prism-shadow
HarborCRM Public-API CRM Handoff Skill

0
installs
▌
Harborcrm JSON Handoff · prism-shadow
Use this skill for HarborCRM public API tasks that ask for CRM-ready JSON handoffs, post-event sponsor or badge reconciliation, trade-show prospecting summaries, or import-batch cleaning. It guides solvers to inspect the prompt and answer template, query only the public HarborCRM API endpoints supplied for the task, normalize and reconcile CRM records, apply exclusions and action conventions, sort deterministically, and return schema-exact JSON.

0
installs
▌
Self Attempt 01 · prism-shadow
MedBridge Sales Ops SOP

0
installs
▌
Medbridge Sales Ops Sop · prism-shadow
Operational SOP for MedBridge Sales Ops quote, RFQ, freight, and milestone reconciliation tasks.

0
installs
▌
Crm Service Ticket Sop · prism-shadow
Use this skill for CRM/support-console tasks that ask Codex to resolve or classify support tickets, mobile service cases, queue snapshots, or enterprise export complaints using customer, account, line, bill, plan, device, outage, diagnostic, SLA, or incident records. Trigger whenever the task says to use a shared support console API and return JSON matching an answer template.

0
installs
▌
Crm Support Ticket Sop · prism-shadow
Use this skill whenever a task asks Codex to resolve CRM, service-ticket, mobile support, contact-center, queue-quality, or enterprise export-complaint cases from payload files plus a shared support-console API. It gives the lookup workflow, strict JSON output conventions, routing and action rules, summary-count habits, and pitfalls for support-console evidence tasks.

0
installs
▌
Apexcloud Retention Ops · prism-shadow
Use this skill for ApexCloud Retention Operations tasks that ask for account retention, churn-risk, support, NPS, revenue, renewal, or pipeline analysis from the remote GDPEVO environment. Trigger whenever the task mentions ApexCloud, accounts, ARR, CSM owners, lifecycle status, product usage, SLA, support tickets, NPS, opportunities, churn candidates, renewals, or asks for a computed business answer from the remote API.

0
installs
▌
Erp Finance Control Review · prism-shadow
Use this skill for task_group_005 ERP finance API tasks, especially reimbursement-to-AP close reviews, stale AP batch refreshes, vendor onboarding or account-change release controls, prepaid close reconciliations, AP aging, payments, vendors, compliance, claims, bills, GL balances, and close-log checks. Trigger whenever the prompt asks for JSON decisions from the shared ERP finance API, even when local payloads look sufficient, because the API is the system of record and the templates have strict field, ordering, and precision rules.

0
installs
▌
Task Group 005 Erp Finance Control · prism-shadow
Solve task_group_005 ERP finance-control evaluations by reconciling current API data for reimbursement AP close, vendor onboarding/account-change release, prepaid close, and stale AP snapshots.

0
installs
▌
Procureops JSON Reconciliation · prism-shadow
Solve ProcureOps procurement, receiving, AP, budget, approval, supplier-risk, and change-control tasks that require schema-exact JSON outputs from the ProcureOps API plus task-local input payloads.

0
installs
▌
Procureops Reconciliation · prism-shadow
Use this skill whenever the user asks for a ProcureOps procurement, receiving, AP invoice, payment-hold, sourcing nomination, contract-change, budget, supplier-risk, or strict JSON decision file based on the ProcureOps API and local input payloads. This is especially important when the task mentions purchase orders, receipts, AP invoices, contracts, requisitions, budget snapshots, approval events, chargebacks, vendor risk, controlled reason codes, or an answer_template.json.

0
installs
▌
Northwind Erp Control Desk · prism-shadow
Use this skill for Northwind Components ERP tasks that require joining API records with a local memo and returning strict JSON decisions for order dispatch, inventory allocation, BOM replenishment, supplier incident scorecards, or supplier quality hold controls. Trigger whenever the user mentions Northwind ERP, shared ERP API, orders, waves, warehouses, inventory, BOMs, purchase orders, incidents, suppliers, allocation, expedite queues, replenishment, or supplier quality review.

0
installs
▌
Northwind Erp Decision Sop · prism-shadow
Use this skill for Northwind Components ERP tasks that require structured JSON decisions from orders, inventory, products, customers, suppliers, incidents, BOMs, purchase orders, transfers, shipping quotes, allocation waves, expedite queues, replenishment plans, or supplier quality scorecards. It gives the reusable API workflow, field conventions, calculation habits, decision precedence, sorting, rounding, and validation checks needed for these tasks.

0
installs
▌
Private Wealth Advisory JSON · prism-shadow
Use this skill for private wealth advisory tasks that ask for a structured planning JSON object for a CLT client using the advisory API, especially Roth conversion/RMD, ILIT Crummey funding, GRAT versus CRAT, or estate liquidity action-plan analyses. Trigger whenever the prompt mentions the private wealth advisory team, CLT client IDs, advisory API records, answer_template.json, RMD factors, life insurance, trust candidates, Roth conversions, ILITs, GRATs, CRATs, or estate liquidity planning.

0
installs
▌
Self Attempt 03 · prism-shadow
Private Wealth Advisory JSON Skill

0
installs
▌
Crescent Finance Ops · prism-shadow
Use this skill for Crescent Arts Collective Finance Ops API tasks involving branch close packages, regional finance views, compensation summaries or forecasts, and weekly payroll reviews. It gives the endpoint workflow, finance/compensation/payroll formulas, rounding, ordering, and JSON answer conventions needed to produce management-reporting outputs from the remote Crescent Finance Ops API.

0
installs
▌
Crescent Finance Ops Reporting · prism-shadow
Use this skill for Crescent Arts Collective Finance Ops API tasks that ask for branch close reporting, regional management views, compensation summaries or forecasts, or weekly payroll reviews. It helps convert request memos, answer templates, and the remote Crescent Finance Ops API into one exact JSON answer with the correct formulas, ordering, rounding, and roster/rate-book edge-case handling.

0
installs
▌
Asteria Portfolio Risk · prism-shadow
Solve Asteria Investment Office finance tasks that require portfolio, credit-risk, equity-correlation, active-allocation, or committee JSON outputs from the shared Asteria API. Use when prompts mention Asteria portfolios, energy-credit tickets, fixed-income rebalance rotations, international equity correlation reviews, macro allocation views, or multi-asset committee files.

0
installs
▌
Peopleops Lifecycle Control Solver · prism-shadow
Use this skill for PeopleOps Console tasks that ask you to reconcile onboarding, leave, payroll, remote-work, recruitment, document, notice, or audit evidence into a strict JSON answer template. It is especially useful when prompts mention source precedence, submitted versus draft records, folder readiness, notice defects, payroll handoff, accrual readiness, or normalized business labels.

0
installs
▌
Cedar Ridge Intake Coordination · prism-shadow
Reconcile Cedar Ridge Intake Coordination Portal data into schema-constrained JSON for patient access verification, referral readiness or chart activation, dialysis transfer packet and capacity review, and chronic-care enrollment panels. Use when a task references Cedar Ridge intake rosters, referral batches, transfer batches, program candidates, patients or charts, ICD metadata, documents, pharmacies, or the portal endpoints.

0
installs
▌
Cedar Ridge Intake Reconciliation · prism-shadow
Reconcile Cedar Ridge Intake Coordination Portal worklists into strict controlled-value JSON. Use when a task references Cedar Ridge roster access verification, referral readiness or chart activation, dialysis transfer packet and capacity review, chronic-care program enrollment panels, answer_template.json payloads, or the shared intake coordination portal.

0
installs
▌
Cedar Ridge Intake Auditor · prism-shadow
Use for Cedar Ridge Intake Coordination Portal tasks that require reconciling patient access, referral readiness, dialysis transfers, chronic-care enrollment panels, chart activation, ICD metadata, insurance/PBM/pharmacy status, packet documents, capacity, and strict template-shaped JSON outputs.

0
installs
▌
Northstar Payer Ops · prism-shadow
Structure Northstar payer-operations outputs for authorization, appeals, claim repricing, peer-to-peer, and finance queue tasks. Use when a prompt references the shared Northstar environment, a target case/claim/appeal/queue ID, a task_context payload, and a required JSON answer template.

0
installs
▌
Skill · prism-shadow bundle
Solve Northstar payer-operations tasks that require reading task_context and answer_template files, querying the shared environment, and returning exact JSON for UM nurse reviews, pharmacy appeals, claim repricing, peer-to-peer summaries, and margin-queue analyses.

0
installs
▌
Ehr Quality Governance Packets · prism-shadow
Build normalized JSON packets and audits from a read-only EHR quality-governance API. Use for duplicate chart merge readiness, referral coordination, care transition handoffs, ServiceRequest quality review, and referral batch audits where Codex must inspect patient, provider, ICD-10, service-code, document, disclosure, encounter, audit-log, and active clinical-list evidence and return schema-conformant JSON only.

0
installs
▌
Ehr Governance JSON Packets · prism-shadow
Build normalized JSON packets and audits from a read-only EHR quality/referral API. Use when Codex must prepare duplicate-chart merge readiness packets, referral coordination packets, care transition summaries, ServiceRequest quality reviews, or referral batch audits from prompt IDs, environment_access.md, and answer_template.json.

0
installs
▌
Ehr Quality Packet JSON · prism-shadow
Generate normalized JSON packets and audits from read-only EHR quality-governance environments. Use when a task asks Codex to inspect EHR/referral/duplicate/service-request evidence through allowed API endpoints and return schema-conformant JSON for duplicate chart review, referral coordination, care transitions, service request quality checks, or referral batch audits.

0
installs
▌
Clinic Protocol JSON · prism-shadow
Produce protocol-bound structured JSON answers from a synthetic clinic runtime. Use when a task asks Codex to review patient, case, observation, medication, allergy, imaging, registry, social-context, or protocol data from a provided clinic environment and return exactly the JSON object specified by an answer template, including evidence identifiers, disposition or routing decisions, safety checks, medication or lab recommendations, red flags, restrictions, and follow-up timing.

0
installs
▌
Synthetic Clinic Protocol JSON · prism-shadow
Produce schema-constrained JSON for synthetic clinic protocol tasks. Use when a prompt asks Codex to review a task runtime with FHIR-like cases, patients, observations, medications, allergies, problems, imaging, care registry, SDOH, or protocol records, then return only a structured clinical decision-support, routing, observation-window, or safety-check JSON object.

0
installs
▌
Investigation Review Hub Analysis · prism-shadow
Use for Investigation Review Hub legal review tasks that require matter-scoped production, preservation, retention, privilege, QC, source-collection, or remediation analysis returned as strict schema-conforming JSON.

0
installs
▌
Investigation Review Hub Remediation · prism-shadow
Use when producing legal investigation, subpoena, SEC, grand jury, retention, production-readiness, gap-analysis, or remediation-dashboard JSON deliverables from an Investigation Review Hub API and a task-provided answer_template schema.

0
installs
▌
Investigation Review Hub JSON · prism-shadow
Structured investigation and eDiscovery review workflow for producing schema-conformant JSON answers from an Investigation Review Hub. Use when a task asks for subpoena, agency, litigation-hold, retention, production-readiness, privilege, QC, collection-gap, or remediation dashboards that rely on task-local payload schemas and hub endpoints as the source of record.

0
installs
▌
Court Closeout Reconciliation · prism-shadow
Reconcile court closeout packets, sentencing or traffic disposition batches, payment-plan orders, and post-sentencing form packets by cross-checking hearing notes, worksheets, petitions, fee schedules, docket entries, and portal form metadata. Use when a prompt asks for a strict JSON answer, corrected disposition or status, fee or payment-plan math, register totals, placeholder handling, or form completion from a court operations portal.

0
installs
▌
Court Packet Reconciliation · prism-shadow
Reconcile court closeout, sentencing, traffic, petition, and post-disposition packet materials into strict schema-matched JSON. Use when a task combines hearing notes, docket entries, case files, intake worksheets, petitions, fee schedules, payment policies, form metadata, or portal records and requires conflict resolution, payment-plan math, unsupported-fee exclusion, placeholder handling for missing identifiers, and ordered structured output.

0
installs
▌
Licensing Structured Review · prism-shadow
Use for structured licensing examiner tasks that require JSON decisions, restricted liquor staff packages, or alcohol renewal manual-review queues from a shared licensing data environment.

0
installs
▌
Licensing Record Review · prism-shadow
Analyze licensing-environment records for schema-strict licensing decisions. Use when a task asks Codex to review contractor application batches, restricted liquor-license transfers or renewals, or alcohol renewal manual-review queues using policy and case-record endpoints, then return JSON decisions, risk codes, verification gaps, monitoring plans, ranked queues, and summaries.

0
installs

997 published skills · page 8 of 10