Atlas Graph Explorer

Wiki Graph Edges

Clusters

catalog-meta1769

CiSurface1
Claim825
ClaimTest5
ClaimTestRun5
DeploymentTarget2
EvidencePolicy2
EvidenceSource227
Gap4
Grammar1
MetaAttribute366
MetaCluster16
MetaEdgeKind84
MetaEnum34
MetaNodeKind114
OntologySchema1
OpenQuestion5
OutOfScopeReason4
PackageSurface23
PathDescriptor45
RunJournalEvent5

extensions1288

domain1003

Domain108
Framework36
Language39
Library63
Platform10
PlatformService65
SkillArea319
Specialization39
StackPart52
StackProfile7
Tool101
Topic164

workflows668

Workflow668

terminology444

Acronym25
Definition116
Synonym11
Term292

agent-stack349

AgentCoreImpl28
AgentPlatformImpl20
AgentProduct38
AgentRuntimeImpl24
AgentUIImpl18
AgentVersion42
CapabilityProfile3
InteractionPrimitive257
InteractionPrimitiveCategory12
LaunchConfig10
Presentation22
SessionModel4

benchmarks232

Benchmark65
EvalHarness5
EvalResult73
EvalRun70
Judge3
Rubric3
TestSet13

role186

Customer3
EndUser3
OrgUnit59
Responsibility49
Role66
Tenant3
VendorRelationship3

sourceref-scope175

ScopeBoundary111
SourceRef64

channels-hooks137

Channel6
DecisionVerb5
HookMapping54
HookSurface70
MergePolicy2

capabilities-and-models133

CapabilitySupport100
Modality7
ModelProviderProduct7
ModelProviderVersion7
ProcessDescriptor2
TransportProtocol10

compute116

APIErrorClass3
AgentHostTransport7
CapacityCascadeSignal2
MCPTransport4
ModelFamily23
ModelTransportProtocol12
ModelVersion49
Provider12
TranscriptIngressEndpoint1
TransportClient2
TransportProxy1

lifecycle106

capabilities85

Capability74
InstallMethod11

wiki44

Page44

security21

FilesystemSafetyInvariant3
HarnessHardeningGuidance3
OutputGuard5
OutputModeChange2
OutputStyle2
PermissionDenialReason3
SecretHandlingPolicy3

stack-layers11

Layer11

context-engineering8

BackgroundConsolidation2
CompactionPolicy2
MemoryHierarchy2
ProactiveSurface2

trust5

TrustLevel5

HomeextensionsFrontmatterField

FrontmatterField

1 of 58 records

Active filters

description: Tool allowlist scoped to this skill's invocation. x clear all

name

description4
name3
model2
id1
kind1
title1
labels1
source1
inputs1
outputs1
io1
metadata1

description

Skill identifier (kebab-case). Required.1
Short prose triggering description used by the model to decide invocation.1
Auxiliary trigger guidance (richer than `description`).1
User-facing arguments hint shown in the slash-command palette.1
Named positional argument schema for substitution.1
When true, skill is user-invokable only — the model cannot auto-trigger.1
When false, skill is hidden from the user palette.1
Tool allowlist scoped to this skill's invocation.1
Pin a specific model version for this skill.1
Pin reasoning-effort level (low/medium/high/xhigh/max).1
Dynamic context injection block (shell commands, file globs).1
Subagent reference to dispatch when this skill is invoked.1

defaultValue

false3
true2
run1
github1
${{ github.token }}1
221
${{ github.workspace }}1
${{ steps.trigger.outputs.matched }}1
${{ steps.trigger.outputs.event }}1
${{ steps.amux.outputs.exit-code }}1

valueType

string29
object11
bool6
enum5
list5
glob2

appliesTo

task-schema16
github-action-input16
skill15
subagent5
cursor-rule3
github-action-output3

required

no49
yes8
recommended1

Sort:id-asc id-desc name-asc name-desc

id	displayName	cluster
frontmatter-field:skill-allowed-tools	Skill frontmatter: allowed-tools	extensions