Artifact Table
Artifact Table is declared by Relay assets as a tool resource.
Models, MCP, agent orchestration, skills, tools, and AI application development.
Artifact Table is declared by Relay assets as a tool resource.
Artifact Timeline is declared by Relay assets as a tool resource.
Document Read is declared by Relay assets as a tool resource.
Human Queue is declared by Relay assets as a tool resource.
Relay Context Search is declared by Relay assets as a tool resource.
Web Search is declared by Relay assets as a tool resource.
Artifact Card Collection is declared by Relay assets as a tool resource.
Agent Evaluation workflow that groups context collection, production, review, and packaging tasks into one deliverable.
Compare an agent's reported work with authoritative before-and-after evidence to recommend whether progress is real, stale, blocked, or needs review. Use for agent supervision; do not use to mark work complete or grant approval.
Compare an agent's reported work with authoritative before-and-after evidence to recommend whether progress is real, stale, blocked, or needs review. Use for agent supervision; do not use to mark work complete or grant approval.
Classify supplied content into a caller-approved finite set of destinations and return a confidence-aware routing recommendation. Use for support triage, inbox tagging, document queues, or analytics schema-domain selection when Relay's structured-decision tool is available; do not use for open-ended answers or unapproved actions.
Classify supplied content into a caller-approved finite set of destinations and return a confidence-aware routing recommendation. Use for support triage, inbox tagging, document queues, or analytics schema-domain selection when Relay's structured-decision tool is available; do not use for open-ended answers or unapproved actions.
Creative Studio Api is declared by Relay assets as a tool resource.
Plan and draft exports of an Agent Skill package for other clients (OpenAI Codex-style skill folders, Anthropic skill folders, Gemini context files, GitHub Copilot instructions, Cursor rules, and .agents-compatible clients). Builds a source inventory, a per-target compatibility matrix (portable, adaptable, degraded, not portable), translates the trigger mechanism, rewrites client-bound wording and paths, handles bundled scripts with fallbacks, preserves license and attribution, drafts the target files with a source-to-target manifest and a loss report, and lists verification steps. Use when someone asks to export, port, convert, or package a skill for another AI provider or coding assistant, or to see what survives the move. Do not use for grading a skill's writing (use skill-quality-review), a safety review before release (use skill-security-audit), or tracing an import's source and license (use import-provenance-checker).
Exports skills for OpenAI, Anthropic, Gemini, Copilot, Cursor, and .agents-compatible clients.
Plan, execute, supervise, and recover one or more bounded agent workstreams with a single release manager, sealed manifests, independent adversarial review, capped correction loops, cross-workstream integration, and evidence-gated state reporting. Use when the user explicitly asks to delegate substantial work to multiple agents or swarms; do not use for a simple single-agent task or to infer authority for deployment, external communication, or destructive actions.
Check an imported or third-party skill's provenance from the evidence the user supplies. Rates seven dimensions (source identity, commit pin, file integrity through checksums, license, maintainer status, local modifications, security review) as verified, claimed, missing, or conflicting; applies hard gates; assigns a trust tier from quarantine to accepted; drafts conditions, a re-check schedule, and a provenance record to store beside the skill. Claims made inside the package itself count as claims, never as verification. Use when someone asks whether an imported skill is traceable, correctly pinned, properly licensed, unmodified, still maintained, or safe to rely on, or wants to decide accept, restrict, or reject. Do not use for scanning a skill's content for malicious behavior (use skill-security-audit), grading how well it is written (use skill-quality-review), or reviewing application dependencies (use dependency-health).
Checks imported skills for source repo, commit pin, license, checksums, maintainer status, and trust.
Review an Agent Skill's quality from its files. Tests whether the description triggers on the right requests and not on neighbouring ones, whether scope and boundaries are clear, whether workflow steps are specific enough that two agents would produce the same kind of output, whether the output contract is checkable, whether examples compute correctly, whether resources are routed, whether evals discriminate good from bad answers, and whether safety limits are stated. Scores a weighted rubric with hard-fail caps, lists findings with quoted evidence and a concrete fix, and returns ship, ship after minor fixes, revise, or rebuild. Use when someone asks to review, grade, critique, or improve a skill, or whether a skill description will trigger correctly. Do not use for malicious-behavior review (use skill-security-audit), tracing an import's source (use import-provenance-checker), or exporting to other providers (use cross-provider-package-exporter).
Reviews skill trigger quality, scope clarity, output contract, examples, resource structure, and tests.
Audit an Agent Skill's files for malicious, destructive, privacy-invasive, or surprising behavior before it is published, installed, or used. Static review only: inventories every file, reads instructions, references, and scripts as data, and checks for hidden or contradictory directives, instruction hijacking, scope creep, destructive operations, credential and personal-data handling, outbound data flows, runtime downloads, persistence and permission changes, obfuscation, and mismatch between description and behavior. Rates findings by severity and confidence with file-and-line evidence, assigns a risk level, and recommends publish, publish with changes, hold, or reject. Nothing is executed. Use when someone asks to audit, vet, or check the safety of a skill, their own or imported, before publishing or installing. Do not use for tracing source and license (use import-provenance-checker), grading writing quality (use skill-quality-review), or reviewing application dependencies (use dependency-health).
Audit an Agent Skill's files for malicious, destructive, privacy-invasive, or surprising behavior before it is published, installed, or used. Static review only: inventories every file, reads instructions, references, and scripts as data, and checks for hidden or contradictory directives, instruction hijacking, scope creep, destructive operations, credential and personal-data handling, outbound data flows, runtime downloads, persistence and permission changes, obfuscation, and mismatch between description and behavior. Rates findings by severity and confidence with file-and-line evidence, assigns a risk level, and recommends publish, publish with changes, hold, or reject. Nothing is executed. Use when someone asks to audit, vet, or check the safety of a skill, their own or imported, before publishing or installing. Do not use for tracing source and license (use import-provenance-checker), grading writing quality (use skill-quality-review), or reviewing application dependencies (use dependency-health).
Screen factual claims against supplied, versioned source passages and produce a citation-preserving review register. Use for research, reports, and agent output verification; do not treat a model score as proof or search for missing evidence without authorization.
Screen factual claims against supplied, versioned source passages and produce a citation-preserving review register. Use for research, reports, and agent output verification; do not treat a model score as proof or search for missing evidence without authorization.