Index of every page on this site: llms.txt
All skills
base avatar

/plugin-review

@ef8bcd5 official
by basebase/skills120 stars
130
Run it yourself

Validate and review Wallet MCP plugin files against the Plugin Specification. Use when writing a new plugin, preparing a plugin PR for submission, self-checking a plugin before opening a PR, or reviewing someone else's plugin PR. Triggers on requests like "check my plugin against the spec", "validate this plugin file", "review my plugin PR", "does this plugin conform to the spec", "prepare my plugin for submission", "write a plugin for X protocol".

Use this Skill: https://skilld.dev/gh/base/skills/plugin-review

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesreport-template.md

≈617 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Evaluation Report Template

Write to a file like PR-<n>-<slug>.md (for PR reviews) or <slug>-review.md (for self-checks). This is the thorough, evidence-based artifact. Be precise; quote the plugin where it matters.

# <slug> — Plugin Spec Evaluation

**Plugin:** `<slug>.md` · **Evaluated against:** current Wallet MCP Plugin Specification · **Date:** <YYYY-MM-DD>

## Verdict
<One of: ✅ Conforms / 🟡 Approve with minor changes / 🔴 Request changes.> <1–2 sentence rationale reflecting both static and (if run) live findings.>

## Conformance Checklist
| Requirement | Status | Notes |
|---|---|---|
| File location `plugins/<slug>.md` | ✅/🔴 | |
| Frontmatter required fields | | title/description/tags/name/version/integration/chains |
| Enum validity | | integration/shell/auth/risk |
| chains ⊆ supported set | | |
| tags (vocabulary, net-new flagged) | | |
| integration most-specific | | |
| `> [!IMPORTANT]` callout first | | |
| `## Overview` | | |
| `## Surface Routing` (+ chat-only behavior) | | |
| `## Orchestration` | | |
| `## Submission` (names tool + mapping) | | |
| `## Example Prompts` | | |
| Conditional sections (Detection/Installation/Auth/Endpoints/Commands/Risks) | | per flags |
| Canonical headings | | |
| Canonical order / anchors intact | | |
| allowlist accuracy | | |
| Neutral language (no yield claims, no steering) | | |
| Geoblock parity (if perps/prediction/gambling) | | frontend vs API |
| Disclaimers in onboarding callout | | |
| High-risk category gate (perps/prediction/privacy) | | legal review needed? |
| Contribution scope (no self-registration) | | SKILL.md / Examples / Conformance table |

## Detailed Findings
<Numbered. Each: severity (blocker/major/minor/nit) — what's wrong — spec ref — concrete fix.>

## Live API / SDK Verification
<Only if run. Table: Endpoint/Command | Tested how | Result ✅/⚠️/❌ | Evidence. + prose noting any contradiction between live behavior and the doc's claims. Update Verdict if live findings change it, noting the change explicitly.>

## Security & Safety Notes
<Specific concerns: untrusted calldata, SSRF, auth/key handling, PII, irreversibility, risk-tag honesty.>

## Recommended Changes
<Concise, prioritized bullets.>

Notes

  • Verdict legend: ✅ conforms · 🟡 approve with changes · 🔴 request changes.
  • If a plugin predates a spec restructure, still evaluate against the current spec and note staleness.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

2 warnings12d3 checks · Risk MEDIUM
  • Gen Agent Trust Hub12d

    This skill is designed to review and validate Wallet MCP plugins by fetching specifications from a trusted repository and using various CLI tools (gh, curl, npm, npx) to evaluate plugins provided in GitHub Pull Requests. A security risk is present in the optional live testing phase, where the agent is instructed to install and execute npm packages specified in the PR content, which could lead to arbitrary code execution if a malicious package name is provided. Additionally, the skill is susceptible to indirect prompt injection as it processes untrusted PR data and external API responses without explicit sanitization.

  • Socket14d

    No alerts

  • Snyk14d

    Risk: MEDIUM · 1 issue

Signed by skilld at ef8bcd5. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 weeks ago.

Activeupdated 2 weeks ago

README badge

README badge for base/skills/plugin-review