New: Skill bounties are live. Post a request, fund the bounty, and creators compete for 7 days to build it -> See open bounties
THE AGENSI STORE
36 skills found
Model quality is table stakes — the harness is where agents win or fail. This designs yours: it writes a structured, testable system prompt (role, tools, boundaries, method, output contract, failure handling) and maps every concern to the right layer — prompt, tool, guardrail, or evaluation — so the pieces reinforce each other instead of fighting.
Designs practical memory architectures for AI assistants, agents, copilots, automations, and workflows, including memory schemas, retention rules, update policies, retrieval keys, summary formats, privacy boundaries, conflict handling, user preference memory, project memory, task memory, and audit-ready memory governance notes for builders.
Your skill works today. Will it work after the next model update? Build the harness that answers with numbers. Builds a standalone regression test harness with mechanical grading to verify skill behavior after model or code updates.
A deep reference library of production agent patterns — orchestration, context, tool design, failure and recovery, oversight, and evaluation. Every pattern states when it applies, when it's the wrong answer, what it costs, and the failure it prevents. Seven reference files, not a checklist.
A defensive catalog of ~39 agent security attack patterns across every surface — injection, tool abuse, exfiltration, memory poisoning, multi-agent trust, retrieval poisoning — each with a detection signal, a concrete defense, and a severity. Nine reference files plus a threat-model worksheet. For hardening agents you own.
Replay automation traces against explicit rules for required steps, order, retries, approvals, idempotency, external side effects, duration, and terminal success.
Turns prompt edits, version notes, testing observations, failure fixes, evaluation outcomes, approval notes, and rollback details into clean prompt change logs, release notes, testing notes, version history, impact summaries, and rollback instructions for prompt-heavy teams, AI workflow owners, agencies, and product teams.
Audits and optimizes checkout and payment experiences for e-commerce stores, SaaS subscriptions, marketplaces, booking platforms, and digital products. Evaluates cart review, pricing transparency, guest checkout, shipping, billing, payment methods, discounts, trust, validation, declines, uncertain payment states, order confirmation, accessibility, analytics, abandoned-checkout recovery, and implementation readiness.
Standardized data retrieval and error handling for the get_data MCP tool.
Turn a research question into a decision-ready report with source grades, claim-to-source mapping, contradictions, and confidence levels.
Diagnose RAG hallucinations, retrieval failures, and citation errors with a structured root-cause audit.
Designs a scored eval suite for an AI agent or SKILL.md skill, measures trigger accuracy, catches regressions against a baseline, and issues a PASS / PASS-WITH-WARNINGS / FAIL release gate.