"Techniques to Improve Agent Programming" matching MCP connectors:
GET /v1/connectors – MCP directory API referenceMatching Connector Tools:
Find relevant evidence. Inspect the sources. Preserve the version your workflow used. Explore all seven Evidence Packs, their included questions and direct free-preview/purchase links: https://varexis.tech/index-packs 1. GitHub + Playwright: Compare repository and browser permissions before connecting an agent to authenticated tools. 2. Notion + Slack: Review workspace access and where content can go before connecting Notion or Slack. 3. Audit & Export Coverage - GitHub, Slack + Cloudflare: Check what a supplier's audit or export claim actually covers before using it in procurement. 4. Next.js Advisory Snapshot: Keep three selected Next.js affected/fixed source assertions and their separately dated enrichment together. 5. Code-Scanner Trust - SonarQube, Snyk & Semgrep: Check scanner permissions and the evidence needed before accepting a clean report or claimed fix. 6. Telemetry Redaction - Coverage, Exceptions & Data Flow: Review where sensitive data can remain or leave a pipeline despite a redaction claim. 7. GitHub Release Acceptance: Check whether approval and execution evidence is bound to the exact artifact being accepted. Complete published packs, sources and limitations are free to inspect and save locally. Public MCP discovery is free, unauthenticated and read-only. These are documentation reviews or selected dated advisory source assertions, not independent tool tests or current safety assessments. Paid value: server-saved retrieval of the selected version. GBP 5 Single Request saves the whole pack; GBP 19 Evidence Receipt also adds an unsigned delivery receipt, plus applicable VAT. Eligible active subscribers retrieve one whole pack using one query without another checkout. No continual refresh or permanent-retention promise. Connect: https://varexis.tech/index-connect Purchase or retrieve: https://varexis.tech/index-access Hosted Stripe and compatible authorised MPP/Stripe SPT purchases use separate authenticated paths. Agents need owner spending authority for the exact VAT-inclusive quote and required provider authentication. Not every agent or wallet is supported. Use saved results or active entitlements before buying. Recover uncertain delivery with the original purchasing path, access key and payment credential; do not buy again. Payment adds no fresh investigation, certification, independent assurance or deployment approval.
Preflight for AI agents, signed: one call; wallet, domain, IP abuse, package checks and CVEs.
Toxicity scores (0-1) for text before an agent posts it. Perspective API compatible. No LLM.
Verify a company before your agent pays, onboards or contracts it. One call returns one verdict (ok / review / blocking / not found) from official public sources: US restricted-party lists (Consolidated Screening List incl. OFAC SDN, refreshed daily), UK Companies House, Canadian federal corporations, GLEIF legal entity status, and contractor licences in Québec (RBQ, RENA), Washington and Oregon. Also: npm/PyPI package safety check. Free tools: qc_contractor_check, qc_licence, pkg_check (daily quota). Paid tools: pay per call in USDC on Base (x402), no account.
Shared second brain for any AI with agent permissions
Check an npm/PyPI package before an AI agent installs it: 5 focused supply-chain tools.
Collision-first MCP diagnostics for MCP, API, webhook and agent failures. Returns structured evidence, bounded repair guidance, agent compatibility checks and safe recovery routing.
Safe to sign? EVM, Solana, XRPL, Algorand txs, permits, tokens; OFAC wallet screen: green/orange/red
Pay-per-call agent tools: PDF, OCR, translation, news search, prediction markets, web, chain reads.
AskYourStack connects Claude to your Linux server and to Google Search Console over MCP. On a server, a small open-source agent (Apache-2.0, reproducible builds) dials out to AskYourStack; no SSH key or password is stored. Claude can then set up a raw VPS, install Magento or WordPress, read logs and fix errors, take backups and snapshots, harden the server, scan for malware and move a site to a new server. It runs as the Linux user who owns the sites by default, or as root when the user chooses
Verify an AI agent artifact: schema, emptiness, provenance and mandate authority, Ed25519-signed.
Independent schema and rule check of AI agent output before you pay. Signed attestation and receipt.
KYA backed by a real company: e-CNPJ-signed agent mandates, revocation-checked, signed verdicts
Prove end users to agents and apps: login-links, OIDC clients, and API keys over remote MCP
Multi-agent coordination outside your LLM context: dedupe parallel work, per-agent budgets, handoff
Verify an artifact an AI agent produced — schema, emptiness, provenance and mandate authority — before you accept the work or pay for it. Ed25519-signed and verifiable without this service; 20 checks a day free, then 0.005 USDC per call over x402.
Review supplied evidence of actor identity, principal authorization, tool scope and intended action before an agent acts. Returns missing evidence and review guidance; enforcement belongs to the caller integration. Two MCP-native free trial attempts per network for the campaign, shared daily capacity. No action authorization or identity certification.
Parse a User-Agent: browser, OS, device, plus bot, search crawler, AI crawler and AI agent flags.
Deploy a repo to a live app from chat: servers, Postgres, domain, email, Stripe, security scan.
Now part of GAIP Witness: verifies GAIP receipts; evidence packs and incident bundles.
