An OpenCode-compatible module that provides securability engineering tools via MCP (Model Context Protocol).
Capabilities:
- Securability engineering review — language-aware FIASSE/SSEM scoring workflow
- Securable code generation — constraint-driven generation contract wrapper
- PRD securability enhancement — requirement-level ASVS + SSEM + FIASSE annotations
- FIASSE reference lookup — topic keyword and section identifier search
opencode.json ← OpenCode configuration (MCP + commands + agents)
instructions.md ← Agent instructions loaded by OpenCode
tools/
mcp_server.js ← MCP server exposing all tools (JSON-RPC 2.0 / stdio)
fiasse_lookup.js ← FIASSE/SSEM reference lookup
securability_review.js ← SSEM scoring and review
secure_generate.js ← Securability-constrained generation contract
prd_securability_enhance.js ← PRD enhancement with ASVS mapping
lib/
common.js ← Shared helpers (zero external deps)
data/
fiasse/ ← FIASSE RFC reference sections
asvs/ ← OWASP ASVS 5.0 requirements
templates/
finding.md ← Security finding output format
report.md ← Assessment report output format
- Node.js 18+
- No external npm dependencies required
Place this module in your project root. OpenCode reads opencode.json and:
- Starts the MCP server — exposing four tools to the agent
- Loads instructions — from
instructions.md - Registers commands —
/securability-review,/secure-generate,/fiasse-lookup,/prd-enhance - Registers agents —
securability-reviewer(read-only analysis agent)
| Tool | Description |
|---|---|
fiasse_lookup |
Look up FIASSE/SSEM sections by topic or section id |
securability_review |
Score code against 9 SSEM attributes across 3 pillars |
secure_generate |
Generate a securability-constrained code generation contract |
prd_securability_enhance |
Enhance PRD features with ASVS + FIASSE/SSEM annotations |
Run these from the OpenCode prompt:
| Command | Description |
|---|---|
/securability-review |
Run SSEM analysis on the current workspace |
/secure-generate |
Run end-to-end securable generation: PRD enhancement, FIASSE-guided implementation, review execution, and final scoring |
/fiasse-lookup |
Look up FIASSE/SSEM reference material |
/prd-enhance |
Annotate a PRD with ASVS + FIASSE/SSEM |
| Agent | Description |
|---|---|
securability-reviewer |
Read-only agent that can analyze but not edit files |
Each tool also accepts JSON from stdin and emits JSON to stdout:
echo '{"topic":"transparency"}' | node tools/fiasse_lookup.js
echo '{"request":"Build a secure file upload handler","language":"Python"}' | node tools/secure_generate.js
echo '{"workspaceRoot":".","targetPath":"."}' | node tools/securability_review.js
echo '{"workspaceRoot":".","prdPath":"docs/prd.md","asvsLevel":2}' | node tools/prd_securability_enhance.js- SSEM scores are heuristic estimates based on static text signals and code shape. Validate with manual engineering review.
- Full semantic code review requires language-aware AST parsers and dynamic analysis beyond what this module provides.
- ASVS requirement mapping uses text-similarity heuristics; requirement-by-requirement traceability requires manual validation.