테스트·보안 스킬

Check a local repository before AI-assisted edits and issue a Green, Yellow, or Red risk receipt. It highlights destructive commands, credential-writing automation, risky agent permissions, and Git state, then recommends whether to proceed, checkpoint, or stop for review.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Check a repository before agent edits
• Review hooks and automation for destructive actions
• Find credential-writing scripts or workflows

Review security in Alchemy applications across credentials, browser access, wallet signing authority, and webhook verification. Map trust boundaries, define data checks, and collect evidence from secret scans and failure tests.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing Alchemy credentials for exposure
• Checking browser access restrictions
• Validating Notify webhook signatures

Review security controls for Adobe integrations, including access permissions, credential rotation, webhook authenticity, signed URLs, and logs. Produce a threat model, access matrix, test evidence, and ownership plan before making changes to live services.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Review permissions in an Adobe integration
• Plan credential rotation with an overlap period
• Test rejection of forged or replayed webhook events

Diagnose failed or degraded Alchemy requests by separating authentication, usage limits, RPC errors, and partial responses. Plan bounded retries, verify recovery and failure behavior, and prepare redacted evidence for escalation.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Investigating Alchemy 429 errors
• Checking partial failures in Portfolio responses
• Separating authentication and request errors

Prepare a bounded support bundle for an Abridge issue using allowlisted operational details such as versions, coarse timestamps, and correlation references. Exclude recordings, transcripts, patient identifiers, credentials, and unrestricted logs, then document redactions and review details.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Preparing reviewed evidence for an Abridge support case
• Documenting redactions and excluded data
• Collecting approved operational details for an incident

Investigate complex Adobe failures layer by layer, from network and authentication to product access, schemas, asynchronous jobs, storage, Runtime, and downstream systems. Compare sanitized evidence with a synthetic control, identify the fault area, and prepare a minimal fix and escalation summary without exposing credentials or content.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Find why an Adobe job is stuck
• Investigate intermittent Adobe API failures
• Separate access problems from malformed requests

Review Adobe integration code and configuration for retired APIs, authentication mistakes, exposed secrets, and fragile workflows. Produce findings with file references, supporting sources, proposed corrections, tests, and unresolved risks before making live changes.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Find retired Adobe API endpoints
• Review authentication and secret handling
• Check asynchronous jobs and signed URLs

Prepare a minimal Adobe diagnostic archive for engineering or support escalation. It uses an explicit collection allowlist, synthetic reproduction evidence, redaction checks, hashes, and custody details so reviewers can inspect what was collected and what remains unresolved.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Preparing a redacted Adobe support bundle
• Reproducing an issue with synthetic test data
• Checking a diagnostic archive for secrets or sensitive content

Review data handling across Adobe workflows, including classification, minimization, storage, signed URL custody, retention, deletion, content provenance, and privacy requests. The skill produces a reviewable record of data flows, decisions, tests, owners, and unresolved gaps.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Mapping data flows in an Adobe workflow
• Reviewing signed URL handling and expiry
• Planning retention, deletion, or legal hold controls

Review security and privacy controls for an Abridge deployment across data handling, identity, devices, EHR workflows, support, audit logs, and incident response. The skill organizes evidence, ownership, gaps, and residual risks without treating a checklist as proof of HIPAA compliance.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing security controls for an Abridge deployment
• Renewing a control assessment with current evidence
• Mapping vendor and health system responsibilities

Build a local test setup for customer-owned Abridge integration code using synthetic data and mocked external calls. Reproduce workflow states and failures, track the interface revision behind test data, and document gaps between simulation and live systems.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Testing integration code without patient data
• Reproducing timeouts and duplicate deliveries locally
• Checking draft review and handoff states

Validate an Abridge pilot through its documented application workflow, from consent and recording to draft note review. Use approved test encounters to check Linked Evidence, safe failure handling, and unintended chart updates, then report go/no-go findings.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Checking core workflows in a new Abridge environment
• Testing consent, recording, and note review
• Sampling Linked Evidence for generated notes

Triage Abridge recording, note review, Epic handoff, and access problems using observed symptoms and approved guidance. Identify the affected workflow stage, organize a minimal timeline, and prepare a redacted escalation record for the responsible team.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Separate recording problems from upload failures
• Investigate note creation delays and editor issues
• Identify the responsible team for Epic handoff failures

Add a synthetic-data CI gate for customer-owned Abridge integration code. Test consent handling, note-review handoffs, redaction, and rollback behavior with mocked vendor boundaries, without connecting CI to a live Abridge or EHR tenant.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Adding regression checks to an Abridge integration repository
• Testing consent and note-release handling
• Checking logs for sensitive data exposure

Build automated checks for Adobe integrations using offline request and response tests. Separate untrusted pull requests from protected live tests, check for exposed secrets and obsolete routes, and produce reviewable release gates and test evidence.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Adding offline tests for Adobe integration changes
• Checking that fork jobs receive no secrets
• Blocking obsolete routes and exposed credentials

Validate OSINT skill metadata and helper scripts, run the content sync check, and perform a smoke test. It helps maintainers check whether the package is ready to use and spot failed checks, including optional network-dependent verification.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Check SKILL.md frontmatter
• Run the OSINT helper scripts
• Verify skill-content synchronization

Discover cloud storage, package supply-chain risks, and cloud-native infrastructure during authorized external reconnaissance. Assess bucket ownership and public access, decode an AWS account ID offline from a leaked key, check dependency confusion signals, and passively fingerprint related services.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Inventory a target's cloud storage exposure
• Decode an AWS account ID from a leaked key
• Check an npm or PyPI dependency-confusion lead

Assess whether a domain’s published DNS policies allow header-From email spoofing, and explain the likely route. The skill also checks SPF lookup-limit failures and dead include targets as supply-chain risks, using passive DNS data only.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Audit a domain’s email spoofing resistance
• Explain why SPF -all alone does not prevent spoofing
• Investigate SPF PermError or long include chains

Map an organization’s internet footprint from its legal identity through corporate relatives, domains, netblocks, ASNs, and live assets. The skill records ownership evidence and confidence, keeps related discoveries out of automatic scanning, and uses paid data sources only as optional enrichment.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Map an organization’s corporate-family footprint
• Find subsidiaries and domains from a legal entity
• Discover owned netblocks and ASNs beyond a seed domain

Run an authorized external OSINT engagement through asset discovery, enrichment, exposure review, and findings synthesis. The workflow verifies key claims and produces a consolidated multi-tab Excel workbook, while reserving active exploitation and out-of-scope assets for operator review.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Run a full external recon on an authorized domain
• Map discovered subdomains and live hosts
• Review exposed content and JavaScript endpoints

Map an organization’s identity provider, tenant boundaries, and federation links, then run authorized pre-auth account-existence checks and build ranked login candidates from names and a confirmed email pattern. The workflow is capped at 20 candidates per tenant and never submits credentials.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Identify an organization’s identity provider and tenant
• Map sibling domains connected through Microsoft federation
• Check account existence with authorized Microsoft or Okta pre-auth endpoints

Run OpenClaw tests and reproduce bugs on remote machines through Crabbox. Choose a suitable execution provider, verify realistic user workflows, and report the actual provider, environment ID, logs, and timing.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Run broad OpenClaw tests in a remote CI environment
• Reproduce installation or Docker bugs
• Verify behavior on macOS, Windows, or WSL2

Verify Skillshare changes in a real Windows guest running in local UTM. Build from a pinned commit, run end-to-end checks with full and basic user tokens, or prepare an isolated hands-on setup, then report results with the commit and architecture.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Test Windows symlink and junction behavior
• Check copy fallback without Developer Mode
• Run Windows end-to-end checks under two user tokens

Run or create end-to-end test runbooks for CLI behavior inside a devcontainer, using isolated ssenv environments. It checks steps and assertions, helps investigate failures, and returns a structured report with pass, fail, and skipped results.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Running an existing E2E runbook
• Validating a CLI workflow end to end
• Testing in an isolated environment