Skip to main content
Glama

cg

Provides a single entry point to a shared memory cognitive graph for writing, reading, reviewing, protecting, evolving, and verifying memories, with status and edge lookup.

Instructions

灵枢认知图唯一入口(38 op:写/读/审/护/演化/元认知/预测/白箱/验证态 status/三元组反查 edges…)。完整 op/参数语义:cg(op="help", query="<op 名 | 参数名 | 工具名>")

Input Schema

NameRequiredDescriptionDefault
aNo
bNo
kNo返回条数(sustain pool_bench…
opYesroute|read|write|verify|review|protect|identity|consistency|metacognition|self_state|evolution|sustain|scrub|predict|causal|forget|goal|task|recent|state_event|info|index_code|index_doc|ref|whitebox|theory|link|session|ingest|export|maintain|consolidate|insight|ccg|status|edges|audit|help
tsNo
capNo
ccgNoCCG 六要素编译器入参:{action, no…
hitNo
idsNo
newNo
oldNo
outNo
pidNoreview decide 的提案 id
refNo
seqNo
setNo
tagNo
bm25No
factNo
goalNo
keepNo
kindNocondition_gap|layer_shift|general
metaNo
modeNolist|show
nameNo
noteNotask op:本轮备注(change 是计划变…
packNo
pathNoindex_code/index_doc 的目录…
peerNo
planNo
roleNouser|assistant|tool-output|command
rootNo
ruleNo
seedNo
slotNo
sortNo取值随 op 变化,完整枚举见 op=help
tagsNo标签(cap:xxx 会作为 route 的建议…
textNorecent op 的事件文本
viewNo
applyNo
batchNo
childNo
cluesNo
depthNo
editsNo
forceNorestore 强制(忽略前置条件:墓碑等)
gatedNo
kindsNo
layerNoanchor|structural|knowledge|contextual|self
limitNogoal/recent 的返回条数;read 的…
poolsNo
pruneNo
queryNoread 的查询
rootsNo
sinceNo
stateNopending|verified|falsified
swarmNo
traitNo
untilNo
writeNo
actionNo取值随 op 变化,完整枚举见 op=help
changeNo
dedupeNo
intentNoroute 的查询意图
issuesNo问题清单(红队打回理由)
markerNo
memoryNosucceeded|failed|blocked|partial
offsetNo
parentNo
prefixNo
reasonNo原因;consolidate contextua…
resultNo
sourceNoconsolidate|verify|manual
windowNo
contentNowrite 的内容(建议含 CCG 5 要素注释…
contextNo当前情境
doc_refNo
dry_runNosustain heal / scrub:只列动…
excludeNo
granteeNo
grantorNo
horizonNo
lexicalNo
messageNo
missingNo
node_idNo节点 id
queriesNo
redteamNopass|reject
sessionNoread(search/recall 分支):会…
subjectNolink evidence:按主体过滤(如 ag…
v_typesNov1|v2|v3
verdictNo取值随 op 变化,完整枚举见 op=help
boundaryNo
categoryNo
deadlineNo
decisionNoaccept|reject|edit|merge
end_timeNo时间算子:查询窗终点(缺省该侧=无界)
entry_idNo
evidenceNoverify 的证据;state_event:出…
max_rowsNo
min_confNo
negativeNo
orderingNo
overrideNo覆盖/删除受保护节点需显式 override=t…
patternsNo
positionNorecord|reflect|verify|output|sustain
prefixesNo
priorityNo
questionNo
relationNosplit_from|extracted_from|merged_from|refined_from|source
sample_nNo
semanticNo
start_idNo
strategyNostratified|risk|random
subjectsNo
validityNo
verdictsNo
aeis_rootNo
auto_healNo
conditionNo
directionNoout|in
entry_idsNo
max_filesNo
max_itemsNo
max_nodesNo
min_deltaNo
min_mergeNo
neighborsNo
questionsNo
skip_dirsNo
statementNo
subsystemNo
time_axisNo时间算子轴:effective(效力轴,read…
timestampNo
acceptanceNo
auto_scrubNo
basis_textNo
conditionsNogoal op:生效条件
depends_onNo
done_limitNo
importanceNo重要性 0-1
max_eventsNo
merge_intoNo
session_idNo
start_timeNo时间算子:查询窗起点(ISO8601/时间戳)。…
subject_idNo取值随 op 变化,完整枚举见 op=help
valid_fromNo双时间轴**起点**(ISO8601,如 202…
action_hintNo
aggregationNoby_relation|by_parent|by_child
auto_evolveNo
bypass_gainNo
consistencyNo
contract_idNo
exclude_docNo检索面排除 doc_ 索引节点(会话日志切片;默…
goal_statusNoactive|done|dropped
incrementalNo
ledger_onlyNo
min_clusterNo
min_jaccardNo
on_conflictNoreject|defer|record
peer_theoryNo
recent_daysNo
sensitivityNo
snapshot_idNo
task_statusNoactive|blocked|done|dropped
valid_untilNo双时间轴**终点**(ISO8601):此刻后不…
window_daysNo
active_limitNo
allow_evolveNo
blindspot_idNo
content_kindNocode|image_desc|text|permission|work_done|work_wip|ccg_marks|hyperedge
derived_fromNo
end_operatorNogt|gte|eq|lte|lt
expand_nodesNo
max_branchesNo
min_severityNoinfo|low|medium|high
peer_versionNo
position_mapNo
sample_limitNo
signers_fileNo
source_layerNo
subject_kindNoself|user|role|agent
target_layerNo
task_runningNo
auto_flywheelNo
beat_intervalNo
budget_tokensNoread 的 token 预算
heal_intervalNo
prune_dry_runNo
actual_node_idNo
check_strengthNohoop|smoking_gun|doubly_decisive
include_recentNo
min_importanceNo
overwrite_liveNo
peer_signatureNo
relation_typesNo
scrub_intervalNo
start_operatorNogt|gte|eq|lte|lt
status_summaryNo
condition_spaceNo条件空间
evolve_intervalNo
importance_hintNo
include_contentNo
include_historyNo
include_partialNo
requested_layerNo
declared_charterNo
excerpt_strategyNo
predicted_node_idNo
require_conditionsNo
verification_basisNo验证基底
non_applicable_conditionsNo不适用条件

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

B3.4/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full behavioral burden. It hints at mutation-related ops such as write, protect, forget, and evolution, but does not disclose permissions, destructive effects, reversibility, protected-node constraints, or return behavior.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The definition is two sentences, front-loaded with the sole-entry-point role and operation families, followed by the help discovery instruction. It is dense but appropriately sized with no wasted text.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 212-parameter multi-op gateway with no output schema and no annotations, the description is very sparse. It points to op=help but does not provide return information, per-op behavior, or enough standalone detail to invoke most operations correctly without an extra help call.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is only 29% across 212 parameters, so the schema cannot fully document itself. The description partially compensates by naming op and query and showing the help lookup pattern, but it does not explain most parameter meanings directly.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description identifies the resource clearly as the sole entry point to the Lingzhu cognitive graph and enumerates 38 operation families including write/read/review/protect/evolution. It does not mention the sibling tool stg, so sibling differentiation is absent, but the core purpose is clear.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives explicit discovery guidance: call cg(op="help", query="<op name | parameter name | tool name>") for complete op and parameter semantics. This is clear context for how to proceed, though it does not state when-not to use the tool or explain alternatives beyond being the sole entry point.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools