wjgoarxiv/lithermes/visual-qa
Claude Code and agent skill by wjgoarxiv/lithermes
- community
Checking that the SKILL.md file is still in the repository.
Hermes-native visual evidence gate.
Install
npx vskill@latest install wjgoarxiv/lithermes/visual-qaSKILL.md
#contract.activation
#contract.inputs
Require a beta contract, material PNG root, hashes, freshness, cleanup, and host provenance when required. Alpha cannot PASS.
#contract.mode_matrix
| Mode | Trigger | Contract | |
|---|---|---|---|
| smoke | Critical surface | Material beta evidence; zero review hashes/receipts | |
| full | Complete inventory | Block without host-owned review provenance | |
| blocked | Capability absent | Emit exact BLOCKED_*; never downgrade |
#contract.procedure
Probe host capture. Validate opened bytes, hashes, freshness, inventory, timeout/cancel, and cleanup. Track per-child re-entry receipts as diagnostics; no combined wait. The parent tracks and merges each accounted receipt.
#contract.outputs
Return evidence or exact blockers: BLOCKED_RENDERER_UNAVAILABLE, BLOCKED_AUTH_UNAVAILABLE, BLOCKED_CAPTURE_PROVENANCE_UNAVAILABLE, BLOCKED_EVIDENCE_STALE, BLOCKED_RENDERER_OWNERSHIP_UNVERIFIED, BLOCKED_INDEPENDENT_REVIEW_UNAVAILABLE. Beta always needs --evidence-root, hash-matched regular PNGs, and a typed host-owned capture receipt unavailable to public JSON input. Alpha is diagnostic-only; full/reference-fidelity also block without review provenance.
#contract.evidence
Canonical litfamily.design-contract/v1beta2; valid litfamily.design-contract/v1beta1 is evidence-eligible compatibility input; litfamily.evidence-manifest/v1beta1 is a separate schema. See references/complete-contract.md.
#contract.hard_stops
No install, profile sharing, auth persistence, cross-repo daemon, or host-config mutation. Self-review cannot PASS.
#contract.anti_patterns
Reject prose PASS, stale images, HTTP-as-render proof, and incomplete cleanup.
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 repositoryCheckingpackages/lithermes-installer/assets/lithermes-plugin/skills/visual-qa/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/wjgoarxiv/lithermes/visual-qa)