Ship agent workflows in 30 seconds. Browse 1,500+ expert-built and security scanned skills. Browse skills
THE AGENSI STORE
101 skills found
Runs an ordered evidence-integrity gate over any AI draft — grade sources, ground claims, verify technical assertions, stress-test — then returns one PASS/REVISE/FAIL ship decision.
An adversarial senior engineer review gate that audits AI-written code for security gaps and logic errors before shipping.
by Nex AI
Canonical Next.js bridge for secure, real-time communication between browser UIs and local agent gateways.
An adversarial reviewer for AI-written code changes. It pressure-tests a pull request or diff for untested branches, silent behavior changes, missing edge cases, over-confident code that only looks right, and weak tests, then returns a PASS / REVISE / BLOCK verdict before the change merges.
Scaffold and audit secure MCP servers with input schemas, confirmation gates, and safety-first tool definitions.
Audit your frontend for accessibility violations before release — flags WCAG failures, gives prioritized fixes, and blocks the broken patterns that get sites sued.
An adversarial security gate to detect and redact PII, secrets, and confidential data before sending prompts.
Production prompts grow by accretion — every failure gets another appended rule until the prompt is two thousand words of contradictions that the model navigates unpredictably
An adversarial gate to verify AI-use disclosures and draft compliant provenance statements for any venue.
Turns Claude into a senior WordPress launch reviewer that audits a site, theme, or plugin against the entire pre-launch standard across 7 weighted domains and returns one objective go/no-go decision with a scored blocker list.
Architect safe, staged code migrations with zero-downtime patterns and automated rollback gates.
An adversarial reviewer for Dockerfiles and container builds. It flags root users, image bloat, unpinned or cache-busting layers, leaked secrets, and missing hardening, then returns a PASS / FIX / BLOCK verdict — before you build or push the image.
Audit your project's dependencies for supply-chain risk before they ship. Detects the ecosystem, runs the right vulnerability scanners against live advisory data, and adds the checks tooling misses — outdated or abandoned packages, typosquatted or suspicious names, risky install scripts, and license conflicts — then returns a prioritized fix list and a PASS / REVIEW / BLOCK verdict. It's npm audit with triage and judgment on top.
Review a database schema, queries, or migration for the mistakes that get expensive in production — bad table design, missing or wrong indexes, slow and N+1 queries, SQL injection, and migrations that lock or break prod. Engine-aware (PostgreSQL, MySQL, SQLite, SQL Server), it runs an ordered review and returns a PASS/REVIEW/BLOCK verdict with prioritized fixes. Schema mistakes are the most expensive kind — this catches them before they ship.
Audit AI-assisted medical and pharma content for compliance-readiness before it enters formal MLR review or journal submission. It checks claim substantiation and on-label scope, reference integrity (the acute AI risk: fabricated or misrepresented citations), fair balance and safety, AI-use disclosure, ICMJE authorship and GPP, COI and funding, data integrity and patient privacy, and adverse-event flags — then returns a PASS / REVISE / BLOCK verdict with the must-fix list. A readiness pre-check built for the regulated reality of medical communications — not a replacement for formal review.
by Corey Jacobs
Run a buyer-readiness check before publishing an AI agent skill package.
A DevSecOps engineer that stands up and tunes static analysis (Semgrep, SonarQube, CodeQL) for high-signal findings — picks the right tool for the stack, writes the config and rulesets, wires a sane CI gate, and tunes out the false positives that get scanners muted.
A professional technical editor's review for your docs. Catches missing context, unclear writing, and unverifiable claims in READMEs, API docs, and changelogs before they ship — with a PASS/REVISE verdict and a prioritized fix list.
Design and write the eval suite for your LLM-powered feature — the metrics that match your failure modes, a golden dataset plan with starter cases, anchored rubrics, LLM-as-judge prompts with the known bias mitigations, and pass/fail gates wired for CI.
Adversarially audit your agent hooks before you trust them. Catches command injection, secret leakage, over-broad matchers, destructive actions, and blocking-logic mistakes in pre/post-tool-use, prompt, and stop hooks — with a PASS or REVISE verdict and severity-ranked fixes.
Write the delegation brief that lets a background or async agent succeed unattended — precise goal, hard constraints, testable acceptance criteria, a verification plan, and stop-and-escalate rules. Turns "go fix the flaky tests" into a spec an agent can actually execute.
by John Barros
Orchestrate human-supervised code migrations across repository fleets with verifier loops and judge review gates.
by John Barros
Architects human-supervised local AI operations with shared memory, SOP playbooks, and strict approval gates.
Review an AI-generated code diff for the failure modes coding agents actually have — claimed-done-but-not-done, gamed or weakened tests, stubs passed off as complete, silent scope creep, hallucinated APIs, and security regressions. Returns an APPROVE or REQUEST CHANGES verdict with a completion check and severity-ranked fixes.