SkillSpector · Issues· 121 open

compare_scan_accuracy: editable-dependency file:// URL resolves to "C:\C:\..." on Windows

AS3 flags current-skill SKILL.md self-references as Agent Snooping

P6 false positive on HTML Output Rules documentation heading

Feature: MCP server registry scanning — complementary to skill scanning

Oversized companion file causes SkillSpector to skip analysis of unrelated skill files (v2.8.2) (another malicious skill example that bypasses this scanner)

Security: prevent host-file disclosure and scanner verdict bypasses

AE3/AE4 analyze binary assets the ledger already marks opaque_content (85-byte PNG -> HIGH, DO_NOT_INSTALL)

No way to scope a scan, so a skill that ships test fixtures cannot pass

## Feature: Role-aware scanning for structured skill formats such as AISOP/AISP

LLM analyzers run with no sampling controls, so the published recommendation is not reproducible

[BUG] Rate-limited providers silently lose analyzer coverage: max_concurrency=10 hardcoded, 429 batches dropped without retry, partial LLM pass not flagged

Add pre-commit hook support and GitHub Actions CI workflow

Coverage is computed but never affects the exit code: a partially-blind scan is indistinguishable from a complete one in CI

P2 "Hidden Instructions" rule produces false positives due to missing word boundary on GET

(feature) Hypothesis & Correctness: a neurosymbolic verification architecture