Skip to content

JPeetz/agent-skills / skills/security/threat-modeling

threat-modeling

Verified

Use this skill when performing security reviews of features, APIs, or architecture changes. Produces a STRIDE threat model document with risk-rated threats, mitigations, and backlog-ready security stories. Use during sprint planning or design review.

FileKindSize
CHANGELOG.mdmarkdown396 B
LICENSEother665 B
SKILL.mdskill md6.4 KB
evals/evals.jsonconfig2.4 KB

evals/evals.json

View on GitHub
[
  {
    "id": "ev001",
    "description": "API threat model — agent produces a STRIDE threat model for a new REST API endpoint",
    "scenario": "A new POST /api/orders endpoint accepts a user's shopping cart and creates an order. The API is behind a gateway with JWT auth. Agent must produce a threat model with data flow, STRIDE analysis, and security stories.",
    "expected_outcome": "Agent identifies data flow (User → Browser || Gateway → OrderService → Database), walks STRIDE, produces at least 3 threats with mitigations and security stories"
  },
  {
    "id": "ev002",
    "description": "SSRF threat identification — agent spots SSRF risk in a webhook handler",
    "scenario": "A feature lets users configure a webhook URL that the server will call on certain events. Agent must threat-model this feature and identify the SSRF risk specifically.",
    "expected_outcome": "Agent identifies 'attacker-controlled URL can target internal services' as an Information Disclosure and/or Tampering threat, recommends URL allowlisting and private IP rejection"
  },
  {
    "id": "ev003",
    "description": "File upload threat model — agent covers upload safety comprehensively",
    "scenario": "A new profile picture upload feature allows JPEG/PNG files up to 5MB. Agent must produce a threat model for the upload flow.",
    "expected_outcome": "Agent identifies threats in at least 3 STRIDE categories: Tampering (malicious file content), DoS (large files, zip bombs), Elevation of Privilege (path traversal), and recommends mitigations for each"
  },
  {
    "id": "ev004",
    "description": "Risk rating consistency — agent rates threats correctly per the rubric",
    "scenario": "Given threats with various likelihood/impact combinations (H/H, H/M, L/L, M/M, etc.), agent must rate risk correctly in each case.",
    "expected_outcome": "Agent correctly rates H+H or H+M as High risk, L+L as Low risk, all others as Medium risk, using the stated rules"
  },
  {
    "id": "ev005",
    "description": "Residual risk documentation — agent properly categorizes Low risks as accepted",
    "scenario": "After completing the STRIDE analysis, there are 2 Low-risk findings. Agent must properly document them as residual risks with acceptance rationale.",
    "expected_outcome": "Agent does NOT create security stories for Low-risk findings. Instead adds them to a 'Residual Risks (accepted)' section with one-line justification per risk"
  }
]
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall