>_Skillful
Need help with advanced AI agent engineering?Contact FirmAdapt
PerryLink

PerryLink

@perrylink

Cordis contributor · DSH community ecosystem builder (not affiliated) · 110k weekly npm downloads · host-compat & plugin-quality tooling

Beijing, China perrylink.github.io/ On GitHub since January 2026

69

Published Tools

1,583

Total Stars

24,229

Weekly Downloads

138

GitHub Followers

437

Public Repos

100/100

Avg Security

Published Tools

14 MCP Servers39 Skills16 Agentsacross 13 categories

PerryLink

Bounded, layered, approval-gated, auditable cross-session memory for DeepSeek Harness (capability seam: ctx.memory + SQLite provider + memory tool + frozen snapshot injection)

...more
Agentdatabase
1461 dir

perrylink

Bounded, layered, approval-gated, auditable cross-session memory for DeepSeek Harness — a capability seam (ctx.memory service + local SQLite provider + memory tool + frozen snapshot injection), not another memory warehouse

...more
SkillLLM Tool
1451 dir

Claude Code-style declarative permission rules for DeepSeek Harness: ordered allow/deny/ask rules with tool-name, argument (glob/regex), and workspace-path matching on the tools/pre-execute waterfall, session-log audit, and HMR reload.

...more
Agentdevtools
1411 dir

Declarative Claude Code-style permission rules plus a Codex-style process-level network policy for DeepSeek Harness: ordered allow/deny/ask rules with tool-name, argument (glob/regex), workspace-path, and network-target (domain/ip/port/scheme) matching on

...more
SkillLLM Tool
1401 dir

perrylink

TypeSafe Jev over the Model Context Protocol: typed judgments (ask, rank, check) that return calibrated probabilities instead of prose.

...more
MCP Serverai
1103 dirs

perrylink

TypeSafe Jev as a first-class Cordis service and three model-visible tools for DeepSeek Harness — offline by default, egress disclosed, never default-on.

...more
MCP Serveruncategorised
1092 dirs

perrylink

Read-only runtime management panel for the official DeepSeek Harness MCP client: /mcp command, a Settings page MCP tab with connection status, recent errors, reconnect counts and tool inventory (sanitized for display), controlled enable/disable patch sugg

...more
MCP Servermcp
983.2K/wk5 dirs

GitHub Actions

The dsh-plugin-guide knowledge base plus the dsh-plugin-dev CLI toolchain: official docs, Cordis primer, community deep-dives, and battle-tested pitfalls registered as an on-demand agent skill, with a scaffolder, static checker, and pack verifier for buil

...more
SkillAI Tool
703.2K/wk1 dir

Delivery quality gate for DeepSeek Harness: grill the requirements, test the implementation, prove the delivery, and gate the handoff with a deliverable/rework decision.

...more
SkillAI Tool
591 dir

Deterministic data profiling, cleaning, and verification for DeepSeek Harness: a ctx.dataQuality capability seam (Service Definition / local Provider / tool Consumers) with data_profile, data_clean, and data_verify model tools, a frozen cross-plugin verif

...more
SkillAI Tool
54991/wk1 dir

LSP action surface for DeepSeek Harness: diagnostics, formatting, completion, code actions, symbols, signature help, inlay hints, and rename tools over language servers

...more
Agentdevtools
421 dir

Cross-device DeepSeek Harness session sync: a dedicated git mirror with append-only keep-both conflict resolution (fork files, never silently overwritten), /sync command and sync_* tools

...more
Agentdevtools
421 dir

LSP action surface for DeepSeek Harness: diagnostics, formatting, completion, code actions, symbols, signature help, inlay hints, and rename tools over language servers, plus the editor action protocol (lsp.actions.*) that makes the plugin the IDE integra

...more
SkillAI Tool
411 dir

Copy Claude Code transcripts, memories, skills and CLAUDE.md into DeepSeek Harness as resumable sessions 鈥?one workspace per project, copy-only, seamlessly resumable

...more
MCP Servermcp
331.4K/wk1 dir

Unified DSH checkpoints: session + workspace + config three-state snapshots with one-shot rollback — /checkpoint and /rewind commands, a checkpoint tool, automatic interval snapshots, a Settings page timeline with pairwise diffs, and seed-replay session r

...more
SkillAI Tool
311 dir

PerryLink

Official-grade GitHub CI for DeepSeek Harness: composite action.yml, PR review bot with idempotent inline comments and a status-check gate, plus PR/issues tools with every write gated by human approval (Apache-2.0, dsh-plugin).

...more
Agentdevtools
311 dir

perrylink

Local-model (Ollama) integration for DeepSeek Harness: discover, pull, remove, and inspect local models, route requests to them by task type or keyword with automatic fallback to the cloud, and get a one-shot status overview via /ollama.

...more
SkillLLM Tool
261.7K/wk1 dir

Provider plugin for dsh-skill-pack-security: registers the pack's skills/ (zh) or skills-en/ (en) edition on ctx.skills AND the plugin_vet supply-chain gate tool on ctx.tools (license/SBOM/commit-lock/malware scans + five-dimension risk card). Ships both

...more
SkillAI Tool
23234/wk1 dir

PerryLink

Voice-first session loop for DeepSeek Harness: a composer microphone button with browser/local speech-to-text (Web Speech, FunASR, whisper.cpp), a speak tool for text-to-speech replies (browser, edge-tts, piper), event announcements with mute, and speak-to-interrupt.

...more
Agentdevtools
221 dir

perrylink

Voice-first session loop for DeepSeek Harness: a composer microphone button with browser/local speech-to-text (Web Speech, FunASR, whisper.cpp), a speak tool for text-to-speech replies (browser, edge-tts, piper), event announcements with mute, and speak-t

...more
SkillSpeech & Audio
21921/wk1 dir

Terminal-style input history for the DeepSeek Harness web composer: edge-first arrows with exact draft/caret restore, browser-local persisted history, Ctrl+R reverse search, workspace-scoped recall, and fully configurable keys — plus sliding-context aware

...more
SkillAI Tool
201 dir

Interactive long-session background agents plus persistent multi-agent team rooms for DeepSeek Harness: durable continuable child agents with progress, steering and interruption, and cross-session team rooms with a message bus, shared task board, approval

...more
AgentAgent Framework
191 dir

PerryLink

PII masking middleware for DeepSeek Harness: anonymize names, phones, emails, ID cards, bank cards, keys, and addresses to placeholders before they reach the model, restore them at the display layer, keep the restore table only in memory and a controlled storage domain, never log plaintext, and expose /mask and the mask_test tool

...more
Agentdevtools
191 dir

PerryLink

OpenTelemetry and Langfuse observability exporter for DeepSeek Harness: turn/step/tool/LLM spans, token and cost metrics, sanitized prompt/completion capture, async batching, bounded offline buffering, retry with backoff

...more
Agentdevtools
171 dir

PerryLink

Unified static-image generation router for DeepSeek Harness: one image_generate tool with standard parameters, config-driven engine routing (OpenAI Images, Zhipu CogView, and any compatible endpoint) with health-aware fallback, durable workspace attachments, per-session quota accounting, and a Plugins settings panel

...more
Agentdevtools
141 dir

GitHub Actions

Multi-dimensional quality scoring for DeepSeek Harness plugins: scores one repo or npm package across install success (consuming dsh-test-drive results), maintenance activity, documentation completeness, security scan, and protocol compliance 鈥?every conc

...more
SkillAI Tool
14365/wk1 dir

perrylink

Unified static-image generation router for DeepSeek Harness: one image_generate tool with standard parameters, config-driven OpenAI-compatible engine routing (OpenAI Images, Zhipu CogView, and any compatible endpoint) with health-aware fallback, durable w

...more
SkillLLM Tool
13929/wk1 dir

PerryLink

Read-only performance diagnostics for DeepSeek Harness: session load/restore timing, spill-hit counts, compaction count and trigger, context-injection volume (AGENTS.md/skills/tool-schema token share), and LLM cache hit rate — surfaced via /fast, persisted as reconstructable session events with async sampling off the model path.

...more
Agentdevtools
131 dir

PerryLink

Laya decision engine as a first-class Cordis service and model-visible tools for DeepSeek Harness.

Agentdevtools
91 dir

perrylink

Laya typed decisions (noul / choice / score) as a first-class Cordis service and model-visible tools for DeepSeek Harness.

...more
SkillAI Tool
8353/wk1 dir

PerryLink

MCP server for Laya typed decisions (noul / choice / score): a warm model sidecar, token-budget preflight, and honest confidence. Node entry point for the Python package.

...more
MCP ServerLLM Tool
56 dirs

PerryLink

TickTick (Dida365) task bridge for DeepSeek Harness: Session-header task panel (list filter, undone/completed views, search, drag reorder), 11 curated agent tools, settings card, over the official TickTick MCP endpoint.

...more
Agentdevtools
51 dir

perrylink

Automatic strong/cheap model-tier routing for DeepSeek Harness: intent-gated tier landing on the agent/request waterfall, plan-mode handoff (strong plans, cheap implements), deterministic high-risk guards on tools/pre-execute, failure escalation with TTL

...more
AgentAI Tool
42 dirs

GitHub PRs, reviews, issues and CI for DeepSeek Harness: approval-gated write tools, a CI composite action, a polling review bot and a status-check gate.

...more
SkillAI Tool
41 dir

GitHub Actions

Persistent multi-agent team rooms for DeepSeek Harness: cross-session rooms with a message bus, a shared task board, approval-gated handoffs and a shared timeline that survive DSH restarts. Extracted from dsh-background-agents 0.9.6, whose background-agen

...more
SkillAgent Framework
41 dir

PerryLink

Vendor parameter translation and deterministic JSON repair for DeepSeek Harness: /translate maps temperature/top_p/max_tokens/stop/system across 11 vendors, and the post-execute repair layer (plus fix_json) fixes broken JSON tool output without ever fabricating data

...more
Agentdevtools
31 dir

https://github.com/PerryLink/dsh-cert-mcp

MCP Serveruncategorised
23 dirs

Temporary package placeholder for staged publishing

MCP Serveruncategorised
23 dirs

perrylink

Temporary package placeholder for staged publishing

MCP Servermcp
22 dirs

GitHub Actions

DSH bundle for Advisers: registers the advisers service and the read-only advisers_brief tool.

SkillAI Tool
21.1K/wk1 dir

GitHub Actions

Token, job and disk budgets with admission decisions and opt-in pre-execute enforcement for DSH agents.

SkillAI Tool
21.6K/wk1 dir

核对施工安全资格台账:三级教育与学时、培训档案四要素、特种作业证书、有效期、工种匹配、六个月实习期。

SkillAI Tool
2540/wk1 dir

perrylink

Shared zero-runtime-dependency toolkit for PerryLink DSH plugins: a pluggable Provider registry seam, fail-closed approval and adaptive session-event gates, mechanical verify scripts, shared sanitize/pricing/judge modules, and a new-plugin skeleton.

...more
SkillAI Tool
21 dir

Goal-driven autonomous dev loop for Claude Code: set a goal, let the 13-phase state machine handle design→impl→test→repair→convergence.

...more
Agentuncategorised
2

perrylink

Read-only MCP server exposing the dsh-plugin-certification registry: certification grades, snapshots and five-dimension evidence for DeepSeek Harness plugins.

...more
MCP Servermcp
1681/wk2 dirs

Read-only MCP server exposing the dsh-plugin-certification registry: certification grades, snapshots and five-dimension evidence for DeepSeek Harness plugins.

...more
MCP Servermcp
1377/wk1 dir

In-app Community Bench tab for the dsh-score public evidence layer: reads the public registry and shows GOE/unrated evidence. No local scoring.

...more
SkillAI Tool
1346/wk1 dir

滞箱费台账核对(按免箱期与费率核对超期天数与金额的自洽性,仅提示差异,不作出定性结论)

Skillai-ml
1112/wk1 dir

按主键对齐两份 Excel/CSV 表,按容差逐字段比较,输出 Markdown 报告、JSON 结果与差异工作簿。

SkillAI Tool
1549/wk1 dir

能力差距与培养计划核对(按能力等级差、培养措施与完成期限核对台账闭环,仅提示差异,不作出定性结论)

SkillAI Tool
1111/wk1 dir

PerryLink contributors

Stdio MCP server for the dsh-os capability catalog: zero dependencies, read-only by default, approval-token write path, no DSH connection.

...more
MCP Servermcp
11 dir

perrylink

Temporary package placeholder for staged publishing

MCP Servermcp
11 dir

项目全流程文件链条核对(按环节先后与金额口径核对链条自洽,仅提示差异,不作出定性结论)

SkillAI Automation
1234/wk1 dir

铁路施工天窗台账核对(按时间窗、时长算术与命令号核对台账自洽,仅提示差异,不作出定性结论)

SkillAI Tool
1255/wk1 dir

土壤与地下水监测台账核对(按点位、监测项目、限值与超标判定核对台账自洽,仅提示差异,不作出定性结论)

SkillAI Tool
1253/wk1 dir

评标因素评分矩阵核对(按分值上限与合计核对矩阵自洽,仅提示差异,不作出定性结论)

SkillAI Tool
1444/wk1 dir

perrylink

Core types and runtime for Advisers: matters, scoped ledgers, focus frames, and the Link SPI.

SkillAI Tool
1.6K/wk1 dir

报关单证一致性核对(按公开的报关单填制规范核对单证间字段一致性,仅提示差异,不作出定性结论)

SkillAI Tool
323/wk1 dir

Dataset documentation for DeepSeek Harness: bounded CSV/TSV/JSON/JSONL profiling (types, missingness, cardinality), redacted PII detection by column name and value pattern (emails, phones, national IDs, IPs), a Datasheet-style card with missing-section wa

...more
SkillAI Tool
252/wk1 dir

perrylink

Public/synthetic eval-pack lint and deterministic JSON grader; not a validated benchmark.

SkillML Testing
224/wk1 dir

Public/synthetic eval-pack lint and deterministic JSON grader; not a validated benchmark.

SkillML Testing
1 dir

perrylink

Regression eval gates for DSH agents: case definitions, baseline comparison and CI verdicts.

SkillML Testing
1.7K/wk1 dir

perrylink

Extract text-layer tables from a PDF and export CSV; no OCR, scanned PDFs are not supported.

SkillDocument Processing
235/wk1 dir

Audit construction progress-payment declarations against the contract baseline and export a review worksheet; amounts in integer cents.

...more
SkillAI Tool
48/wk1 dir

Shared zero-runtime-dependency toolkit for PerryLink DSH plugins: a pluggable Provider registry seam, fail-closed approval and adaptive session-event gates, mechanical verify scripts, shared sanitize/pricing/judge modules, and a new-plugin skeleton.

...more
SkillAI Tool
1 dir

MCP management console for the official DeepSeek Harness MCP client: /mcp command with health diagnostics and pipeline trial calls, a Settings MCP tab with server CRUD (append-only profile-patch fragments, approval-gated writes with automatic backups), a

...more
MCP Servermcp
1 dir

GitHub Actions

Data-management-plan tooling for DeepSeek Harness: a ten-section scaffold with prompts, completeness/consistency checks (datasets, personal data, open-sharing conflicts, retention minimum, backup, repository/license/DOI, contact), Markdown/JSON/SARIF expo

...more
SkillAI Tool
1 dir

GitHub Actions

中文基金申请书检查(DeepSeek Harness):19 项起步清单(项目名称/摘要/科学问题属性 A-D/立项依据/目标/内容/方案/技术路线/创新/可行性/基础/条件/团队/预算/预算说明/伦理/数据共享/限项/参考文献),预算算术与结构校验(设备费/业务费/劳务费/间接费用,合计一致性),申请材料包扫描(申请书/预算表/伦理/承诺书/简历/附件),输出 JSON/Markdown 与 REP 证据。关键词清单,不是评审结论。

...more
SkillAI Tool
1 dir

perrylink

Name-reservation placeholder for layacore-mcp (a Python MCP server exposing Laya typed decisions). The Node entry point is a non-functional shim. Work in progress.

...more
MCP Servermcp
1 dir