Proposal: add a DeepSeek Harness (DSH) adapter — purely additive under dsh/
Nobody has claimed this yet.
Assessment
- Difficulty
- 5/5
- Estimated time
- Over a week
- Newbie friendliness
- 30/100
Research direction
Start by reviewing the proposed dsh/** layout, including dsh/skills/, dsh/setup.md, dsh/config/mcp-dsh.patch.example.yml, and dsh/scheduler/. The first decision is whether this adapter belongs as a purely additive contribution here or in a standalone repository; done means that scope is accepted and the contribution path is clear.
Written by the indexing model from the issue text.
Description
Summary
I ported LLMQuant Data's orchestration skills to DeepSeek Harness (DSH) — the 6 playbooks in skills/ (morning-brief, earnings-watch, 13f-diff, portfolio-pulse, polymarket-watch, llmquant-setup) now also run under DSH via the @llmquant/data-mcp server. I'd like to contribute this back.
Before opening a PR I want to check scope: would you accept an additive dsh/ adapter inside this repo, or would you prefer it as a standalone repo under LLMQuant?
What the adapter contains
dsh/skills/— the 6 playbooks rewritten for DSH:- tool names use DSH's
mcp__llmquant-data__<tool>prefix; hermes cron/--deliver telegramreplaced with DSH's actual scheduler (dsh-schedulesession-local reminders, plusdsh --profile headlessas the OS-cron primitive);- Hermes persona/config references (
~/.hermes/SOUL.md,hermes skills tap add) replaced with DSH equivalents (AGENTS.md,<workspace>/.dsh/skills).
- tool names use DSH's
dsh/setup.md— DSH wiring:cordis.patch.ymlinsert via@deepseek-ai/dsh-mcp-client.dsh/config/mcp-dsh.patch.example.yml— config template (API key masked).dsh/scheduler/—dsh headlesswrapper + Windows Task Scheduler / cron examples (DSH has no built-in calendar cron).
Guarantees
- Purely additive: the PR only adds
dsh/**. Existingskills/,templates/,config/, andSETUP.mdremain byte-identical — no change to any Hermes behavior. - Attribution kept: it's a derivative of this repo (and
LLMQuant/skills), both MIT; the notices stay in place.
Why this helps
DeepSeek Harness ships an MCP client, skill discovery, and a headless CLI, but there's no upstream LLMQuant path for it. The 18 analysis skills in LLMQuant/skills are framework-agnostic and reused unchanged; only these 6 orchestration playbooks needed porting.
If you'd rather keep this repo Hermes-only, just say so — I'll publish the adapter as a separate repo and open a small README cross-link instead.
- Dominant language
- No language data
- Stars
- 26
- Forks
- 7
- PR merge metrics
- No merged PRs in 30d
Getting set up
This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from LLMQuant/llmquant-hermes
-
Difficulty 5/5 Over a week Newbie friendliness 25/100
All issues in LLMQuant/llmquant-hermes
Similar issues
-
Difficulty 2/5 1-2 days Newbie friendliness 68/100
Maintainers usually reply within 1 day
-
bug good first issue help wanted Project:Tool workshop
Difficulty 2/5 1-3 hours Newbie friendliness 90/100
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 82/100
Maintainers usually reply within 1 day
-
bug good first issue help wanted
Difficulty 2/5 1-3 hours Newbie friendliness 90/100
ysb3996-blip/oops#24 ·
Maintainers usually reply within 1 day
-
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
lognorman20/monaco#7428 ·