When and how to reach for the companion detectors -- bandit (Python SAST) and trivy (deps + secrets + IaC misconfig) -- alongside the core semgrep/CodeQL/osv/trufflehog toolchain
Scanned 8/31/2026
Install via CLI
openskills install deonmenezes/mantishack---
name: detection-breadth
description: When and how to reach for the companion detectors -- bandit (Python SAST) and trivy (deps + secrets + IaC misconfig) -- alongside the core semgrep/CodeQL/osv/trufflehog toolchain
---
The core Detect toolchain is semgrep (broad SAST), CodeQL (dataflow SAST), osv-scanner (SCA), trufflehog (secrets), and program-analysis (AST/taint/SMT). Two companion servers widen coverage; reach for them deliberately, not reflexively.
- **`bandit_scan`** (mantis_bandit): Python-specific SAST. Use it on Python targets in addition to semgrep -- bandit encodes Python-idiom checks (e.g. `subprocess` with `shell=True`, weak crypto, `yaml.load`, flask debug) that a generic ruleset can under-cover. Every hit is a `candidate`; bandit's own severity/confidence describe rule confidence, not demonstrated impact, so do not report them as final severity. Default `confidence` to `low` for full recall on a first sweep.
- **`trivy_scan`** (mantis_trivy): composition analysis in one pass -- vulnerable dependencies, embedded secrets, and IaC (Dockerfile/Terraform/k8s) misconfigurations. Use it to add container/IaC coverage that osv-scanner (deps only) and trufflehog (secrets only) don't reach. A vulnerable dependency being present does not mean its vulnerable code path is reachable -- that's a separate question for the reachability stage. Secret values are never returned, only rule/line references.
Both degrade gracefully: if the underlying binary isn't installed they report `available: false`. When that happens, say so explicitly rather than claiming coverage you didn't get, and fall back to the core toolchain. Do not double-count: if trivy and osv-scanner both flag the same advisory for the same package, it's one candidate, not two.
Every result from either tool is a `candidate` in the findings lifecycle -- register it via `mantis_findings` `finding_create`, then trace reachability and attacker-simulate before it can become `confirmed` (see the `mantis-pipeline` and `findings-spine` skills).
No comments yet. Be the first to comment!
Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...
**Complete production-ready guide for Google Gemini embeddings API** This skill provides comprehensive coverage of the `gemini-embedding-001` model for generating text embeddings, including SDK usage, REST API patterns, batch processing, RAG integration with Cloudflare Vectorize, and advanced use cases like semantic search and document clustering. ---
Interview, source-challenge, verify, save, and ADR-gate fuzzy coding requests into Codex-ready implementation specs. Use when a feature, bugfix, refactor, migration, repo-wide change, or architecture task needs user-verified requirements, source-backed decisions, durable architecture decisions, acceptance criteria, validation commands, rollout notes, saved spec/ADR files, and a Codex execution prompt. Do not use when already fully specified or when the user wants direct implementation now.
Use when a repo needs CodeGraph plus ast-grep for Codex MCP setup, exploration, impact analysis, structural search, or safe refactor planning.