AI agent tool reliability // full index

Tools

Shell @livx.cc/agentx — Execute a real /bin/sh command (CLI disk mode) or a sandboxed bash-like interpreter over the VFS (library/sandbox mode).
Read @livx.cc/agentx — Read file contents with 1-indexed numbered lines, optional offset/limit.
Grep @livx.cc/agentx — Search file contents using a pattern.
Glob @livx.cc/agentx — List files matching a glob pattern.
Write @livx.cc/agentx — Write content to a file.
MultiEdit @livx.cc/agentx — Apply multiple edits in one call.
TodoWrite @livx.cc/agentx — Write to a planning scratchpad.
Task @livx.cc/agentx — Spawn a depth-limited child agent over the VFS.
SlashCommand @livx.cc/agentx — Run a reusable prompt template from a commands directory.
WebFetch @livx.cc/agentx — Fetch a URL as readable text.
WebSearch @livx.cc/agentx — Search the web (keyless DuckDuckGo by default, upgrades to Firecrawl/Tavily if key set).
WebSearchAnthropic @livx.cc/agentx — Anthropic's native web search relayed through a cheap model.
get_component reactbits-dev-mcp-server — Get the source code for a specific ReactBits component.
search_components reactbits-dev-mcp-server — Search for ReactBits components by name or description.
get_component_demo reactbits-dev-mcp-server — Get usage example and demo code for a ReactBits component.
list_categories reactbits-dev-mcp-server — List all available component categories.
list_components reactbits-dev-mcp-server — List all available ReactBits components with optional filtering.
intake-init atlasbound — Install the intake + resume/handoff prompts into .claude/.
setup atlasbound — Detect stack, scaffold .atlasbound, build, wire hooks, and produce L1 verdict.
mcp__atlasbound__setup atlasbound — Scaffold AtlasBound in the current repo (detect stack, create .atlasbound, build, wire hooks, L1 verdict).
mcp__atlasbound__update atlasbound — Refresh the code-intelligence snapshot after code changes (honest no-op when already fresh).
mcp__atlasbound__intake atlasbound — Run the mandatory pre-work interview and scope lock for a development task.
mcp__atlasbound__handoff atlasbound — Create a bounded, parent-linked development handoff pack for continuity across /clear.
onboard atlasbound — Run the live-measured checklist, do exactly NEXT, call again, repeat to 8/8.
update atlasbound — Refresh the snapshot after code changes (honest no-op when already fresh).
analyze_files mcp-gemini — Analyze multiple files with Gemini CLI
analyze_directory mcp-gemini — Analyze directories with Gemini CLI
verify_feature mcp-gemini — Verify feature implementation in codebase
raw_prompt mcp-gemini — Execute raw prompts without file context
get_config mcp-gemini — Configuration management - get config
set_config mcp-gemini — Configuration management - set config
optimize_error_message @realpkuasule/harness-automation — 错误信息优化和建议
collect_ab_metrics @realpkuasule/harness-automation — 收集 A/B 测试数据
analyze_ab_results @realpkuasule/harness-automation — 分析 A/B 测试结果
query_state @realpkuasule/harness-automation — 查询当前进度（支持断点续做）
evaluate_rules @realpkuasule/harness-automation — 评估并推荐规则（四问题判定流）
scan_codebase @realpkuasule/harness-automation — 扫描代码库发现违规模式
generate_config @realpkuasule/harness-automation — 生成配置文件（CLAUDE.md, ESLint, Husky, CI 等）
confirm_decisions @realpkuasule/harness-automation — 确认规则推荐结果
rollback @realpkuasule/harness-automation — 回滚到备份状态
cognitive_skill @realpkuasule/harness-automation — 认知层技能（诊断 / 教育 / 决策支持）
get_rule_stats @realpkuasule/harness-automation — 规则效果统计
export_rules @realpkuasule/harness-automation — 规则跨项目复用（导出）
import_rules @realpkuasule/harness-automation — 规则跨项目复用（导入）
list_rule_presets @realpkuasule/harness-automation — 预设模板列表
list_rule_exports @realpkuasule/harness-automation — 导出列表
reset_state @realpkuasule/harness-automation — 重置状态机
validate_setup @realpkuasule/harness-automation — 验证配置完整性和语法正确性
start_ab_test @realpkuasule/harness-automation — 启动 A/B 测试对比介质效果
analyze_rule_adjustments @realpkuasule/harness-automation — 自适应调整建议
suggest_error_improvement @realpkuasule/harness-automation — 错误信息模板效果评估
init_harness @realpkuasule/harness-automation — 一键初始化（v1.2.0: 支持 gitProvider + collaborationMode 参数）
assess_suitability @realpkuasule/harness-automation — 评估项目是否适合应用约束体系
hablala_update_policy @hablala/mcp — Updates an existing policy.
hablala_delete_policy @hablala/mcp — Deletes a policy. Destructive.
hablala_create_record @hablala/mcp — Creates a new record in an object.
hablala_update_record @hablala/mcp — Updates an existing record.
hablala_delete_edge @hablala/mcp — Deletes an edge. Destructive.
hablala_describe_objects @hablala/mcp — Lists all objects in the workspace dictionary with optional verbosity control.
hablala_describe_object @hablala/mcp — Describes a single object by its slug.
hablala_search_objects @hablala/mcp — Searches objects by text in slug, title, description, or synonyms.
hablala_query_records @hablala/mcp — Queries records of an object with optional filters and verbosity.
hablala_get_record @hablala/mcp — Reads a single record by its ID.
hablala_list_policies @hablala/mcp — Lists all policies in the workspace.
hablala_create_object @hablala/mcp — Creates a new object in the dictionary.
hablala_delete_record @hablala/mcp — Deletes a record. Destructive.
hablala_restore_record @hablala/mcp — Restores a deleted record.
hablala_create_edge @hablala/mcp — Creates an edge in the graph.
hablala_update_object @hablala/mcp — Updates an existing object's metadata.
hablala_update_attribute @hablala/mcp — Updates an existing attribute.
hablala_archive_attribute @hablala/mcp — Archives an attribute. Destructive.
hablala_create_relationship @hablala/mcp — Creates a relationship between two objects.
hablala_archive_object @hablala/mcp — Archives (soft-deletes) an object. Destructive.
hablala_add_attribute @hablala/mcp — Adds an attribute to an object.
hablala_create_policy @hablala/mcp — Creates a policy (published cut).
qa_tabs qiksy-mcp — Every browser tab Qiksy can read, incl. which are isolated multi-login sessions (isolatedSession = the login name). Use a tab's tabId with t
qa_findings qiksy-mcp — Findings (with selectors + detail); severity = all | error | warning.
qa_export qiksy-mcp — The full qa-export/v1 bundle: findings, failed requests + bodies, form structure, repro steps, env.
qa_snapshot qiksy-mcp — A full-page accessibility snapshot — interactive elements + landmarks with their roles, labels and a ref you can pass to qa_click/qa_type/qa
qa_tour qiksy-mcp — Return the generated exploratory-testing checklist (from forms + interactive surfaces), auto-checked from the recorded session — raw materia
qa_report qiksy-mcp — Generate Qiksy's self-contained HTML session report and return the HTML (needs History recording to have captured steps).
qa_spotlight qiksy-mcp — Dim + outline an element by CSS selector so the human can see what the agent means. A non-destructive overlay — no click/focus/mutation.
qa_focus_tab qiksy-mcp — Bring a tab in front of the human — activate it, raise its window — so a walkthrough happens where they are actually looking. The one verb t
qa_status qiksy-mcp — URL + error/warning/form counts + isolated-login name.
qa_open_panel qiksy-mcp — Open/close the Qiksy side panel and optionally jump to a tab (forms | findings | tour | tools | history | roles).
qa_run_audit qiksy-mcp — Run the a11y/markup audit on the current DOM and return the findings (also refreshes the panel). Static, read-only DOM analysis.
docbrain_suggest_capture docbrain-mcp — Suggests a capture for files involved in a resolved error or non-obvious behavior, checking if a gap exists in the org's knowledge.
docbrain_capture docbrain-mcp — Drafts a 3-5 line capture of the resolved issue and files it into the review queue after user approval.
docbrain_annotate docbrain-mcp — Drafts a 3-5 line capture for a suggested gap and files it into the review queue after user approval.
get_catalog_item @anychat-ai/mcp — Get a catalog item
get_identity @anychat-ai/mcp — Get agent identity
update_identity @anychat-ai/mcp — Update agent identity
get_personality @anychat-ai/mcp — Get agent personality
set_personality @anychat-ai/mcp — Set agent personality
import_personality_template @anychat-ai/mcp — Import personality template
get_objectives @anychat-ai/mcp — Get agent objectives
set_objectives @anychat-ai/mcp — Set agent objectives
import_objective_template @anychat-ai/mcp — Import objective template
get_guardrails @anychat-ai/mcp — Get agent guardrails
set_guardrails @anychat-ai/mcp — Set agent guardrails

AI agent tool reliability // full index

Categories

Tools