dbillion/hermes-skills-public/cli-capability-audit
Claude Code and agent skill by dbillion/hermes-skills-public
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Verify CLI auth and real commands before claiming value.
Install
npx vskill@latest install dbillion/hermes-skills-public/cli-capability-auditSKILL.md
The same SKILL.md was published earlier as dbillion/hermes-skills-public/cli-capability-audit.
CLI Capability Audit
When to use
- User: "i have X cli", "what value can you bring with X", "what can you do with X".
- Onboarding a new CLI into the workflow, or about to use a CLI unverified this session.
Process (always verify, never assume)
- Confirm binary exists —
which <cli>/lscommon paths. Note exact path (some live in nvm/uv/venv, not default PATH — e.g./home/deeone/.local/bin,/home/deeone/.nvm/.../bin). - Verify AUTH with a real read-gated call — NOT
--help:
- gws →
gws drive files listorgws gmail users getProfile --params '{"userId":"me"}' - nlm →
nlm notebook list - kaggle →
kaggle config view+kaggle competitions list - hf →
hf auth whoami(NOThf whoami— that command doesn't exist) - colab →
colab log
If real data returns → authed. If "Not logged in"/auth error → flag clearly, give exact login command; do NOT describe write capabilities you can't use.
- Enumerate REAL commands —
<cli> --help. Capture actual subcommands; don't invent. - Propose 2–3 concrete tasks from the user's actual data (their notebooks, Drive files, repos). Tie to ongoing work.
- Flag honest limits — plan restrictions, missing auth, read-only scope, what the CLI CANNOT do (e.g. NotebookLM/gws won't post to LinkedIn; agent-reach has no publish command).
Pitfalls
- Never describe capabilities from memory alone. A live read call is the only proof.
- Don't fabricate commands. Gotchas:
kaggle whoami→kaggle config view;hf whoami→hf auth whoami;nlm notebooks list→nlm notebook list. - Don't guess repo/dataset slugs. Use web-verified real repos. HF public-domain books:
manu/project_gutenberg,common-pile/project_gutenberg,stevez80/Sci-Fi-Books-gutenberg. Large parquet downloads can exceed 60s timeout — that's a timeout, not a failure. - Chain value across the stack: HF book → Colab process → NotebookLM summarize → gws store.
This-user stack (verified this session — re-verify each session)
- gws (Google Workspace) → <YOUR_EMAIL>, authed. 17 services.
- nlm (NotebookLM) → 22 profiles, notebooks listed. 2026 upgrades: research-from-idea (
nlm research start/status/import), studio/code-exec artifacts (nlm studio), export to Docs/Sheets (nlm export to-docs/to-sheets), audio overviews (nlm audio). - kaggle → oludayoadeoye777, authed. v2.0.0 (upgrade to 2.2.2 available).
- hf →
/home/deeone/.local/bin/hf, NOT logged in. - colab → 9 named sessions (dsa-nb-*), authed.
colab lsshows no active VM (history only);colab newstarts one. - agent-reach + opencli → 5 social platforms, LinkedIn MCP persistent (systemd). (agent-reach skill is user-owned — do not patch; recommend
hermes curator adoptif fixes needed.)
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/productivity/cli-capability-audit/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/dbillion/hermes-skills-public/cli-capability-audit)