dsh-plugin-development@1.11.0
VERIFIED· install onlyPublisher claims: "A development-time skill for building plugins on DSH (DeepSeek Harness) / Cordis. It is not itself an installable DSH plugin; it is the tooling that guides a developer or AI agent through creating, debugging, packaging, and publishing one. This skill should be used when the user asks to develop a DSH plugin, write or modify a cordis plugin, add a tool, slash command, config schema, service, UI slot or HTTP route to DSH, run a periodic task inside a plugin, fix a plugin that fails to load or stays in PENDING, or package a plugin bundle for installation or distribution. It covers the plugin forms and official conventions, with 12 templates, 28 catalogued pitfalls, and copy-ready prompts for AI-agent pair development. Because DSH is pre-stable and ships breaking changes on a short cadence, it also ships a mandatory pre-coding version gate, a probe that re-verifies its own claims against a live source checkout, and a per-tag history of upstream breaking changes." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Bundled scripts ran with no arguments; no agent task scenario for this bundle.- Tested
- 8 of 10 bundled scripts run with no arguments (the rest were not run: per-bundle cap), with decoy credentials in the environment · every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
What changed since 1.8.1
vet-skills diff dsh-plugin-development 1.8.1 1.11.0This version reaches for something 1.8.1 did not: 1 new host (ghfast.top). Compared from the two recorded runs, not from source.
- New hosts
- ghfast.top
The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-18 and scored it 88/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 1 s · 2 scenarios · 0 deviationsStatic findings
out = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.# ⚠️ 版本号必须写「精确版本」。npx 会去 registry 取包并直接执行,Fix: Pin versions and sources; vendor anything fetched at install time.# bash dsh-sync.sh --mirror <前缀> # 走镜像,例如 --mirror https://ghfast.top/Fix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.command: npxFix: Pin versions and sources; vendor anything fetched at install time.bash dsh-sync.sh --mirror https://ghfast.top/ --check-onlyFix: Replace encoded or assembled strings with plain literals and remove undeclared network calls.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.