threat-model-triage
Triage one inbound vulnerability report, scanner hit, fuzzer artifact, or AI finding against a finished threat model. USE WHEN asked whether a finding is valid or in scope. Applies the §1.1 routing algorithm and §1.17 precedence to assign exactly one closed disposition with section and provenance citations; inferred claims may escalate but never close, and assumptions close only what the model's declared triage policy permits. A finding with no supported route is MODEL-GAP and triggers §1.16. DO NOT USE FOR: producing a model from scratch or backtesting a corpus.
alpha-omega-security · Static checks on file
Installation
Read the source and dependencies before running a command. Copying does not install or execute anything.
npx skills add alpha-omega-security/threat-model --skill threat-model-triageAbout this skill
name: threat-model-triage description: >- Triage one inbound vulnerability report, scanner hit, fuzzer artifact, or AI finding against a finished threat model. USE WHEN asked whether a finding is valid or in scope. Applies the §1.1 routing algorithm and §1.17 precedence to assign exactly one closed disposition with section and provenance citations; inferred claims may escalate but never close, and assumptions close only what the model's declared triage policy permits. A finding with no supported route is MODEL-GAP and triggers §1.16. DO NOT USE FOR: producing a model from scratch or backtesting a corpus. argument-hint: ' + '
Threat Model — Triage (route one finding)
The consumer side of the model. Given an inbound finding and a finished (or drafted) threat model, assign exactly one disposition from the closed §1.17 set, citing the section that licenses it. The response is "see threat model §X", not ad-hoc prose.
Requires a threat model produced by the threat-model orchestrator (or an
equivalent document following
output-structure.md). If no
such model exists, stop and recommend producing one first.
Route from the prose and the threat-model.yaml sidecar, never from the
threat-model.json export. The JSON drops disposition_precedence,
disclaimed-property tiers, and known-non-finding component/symptom scope, so a
consumer holding only the JSON cannot apply §1.17 first-match ordering or the
security-critical floor. It states the contract; it cannot route a report
against it.
Routing algorithm (the §1.1 triager quick-start)
- Locate the sink → look up its row in the §1.7 input-trust table (or the §1.8 output statement, for findings about what downstream consumers may assume).
- Locate the contract dimension → for state corruption, overflow, recursion, callback execution, deserialization, lifecycle, concurrency, or complexity findings, consult the component's contract-dimension row and its routed owning claim in §1.3/§1.5/§1.7/§1.10/§1.11/§1.12 or unresolved question in §1.18. Do not infer failure semantics from the parameter trust row alone.
- Check the attacker capability and control kind the finding requires against §1.7/§1.10. Distinguish control of data from control of size, type/class, callback code, object topology, collaborator implementation, or serialized state.
- Check the affected component against §1.2/§1.3, and any required build flag against §1.6.
- If the root cause lies in a dependency, apply §1.9.
- Check §1.15 for an exact known non-finding match: same component or sink, symptom/attack class, and required conditions, with a current stable discharge reference. Textual resemblance is insufficient.
- Apply §1.17's first-match precedence and assign exactly one disposition,
citing the licensing section and provenance. If none
fits, assign
MODEL-GAPand trigger §1.16 — do not improvise.
The closed disposition set (§1.17)
| Disposition | When | Licensed by |
|---|---|---|
VALID | Violates a claimed property via an in-scope adversary and input. | §1.11, §1.7, §1.10 |
VALID-HARDENING | No §1.11 property violated, but a §1.14 misuse is easy enough to harden. Private; maintainer discretion; usually no CVE. | §1.14 |
OUT-OF-MODEL: trusted-input | Needs attacker control of a parameter marked trusted. | §1.7 |
OUT-OF-MODEL: adversary-not-in-scope | Needs an excluded attacker capability. | §1.10 |
OUT-OF-MODEL: unsupported-component | Lands in out-of-scope code. | §1.3 |
OUT-OF-MODEL: non-default-build | Requires a configuration explicitly marked dev-only, discouraged for the modeled use, or unsupported; non-default alone is insufficient. | §1.6 |
OUT-OF-MODEL: dependency-contract | Root cause is a dependency failing its own contract; usage conformant. Forward upstream, don't just close. | §1.9 |
BY-DESIGN: property-disclaimed | Concerns a property explicitly not provided. | §1.12 |
KNOWN-NON-FINDING | Matches a documented recurring false positive. | §1.15 |
MODEL-GAP | Fits none of the above → revise the model. | triggers §1.16 |
Guardrails
- Exactly one disposition. Apply §1.17's precedence when multiple
preconditions fail. Report
MODEL-GAPonly when the model is silent or genuinely contradictory, not merely because two predicates are true. MODEL-GAPis not "other". It means the model is incomplete: the correct response is to add the property to §1.11/§1.12 (viathreat-model-authoring), not to make an ad-hoc call on the report.- Closure constraint (all statuses), governed by the declared triage policy.
Any disposition that closes a report against the reporter (
OUT-OF-MODEL: *,BY-DESIGN: *,KNOWN-NON-FINDING) must be licensed by a documented or maintainer claim, with one policy-scoped exception for assumption:- An inferred licensing claim always escalates instead of closing.
- Read the header's triage policy. Under
strict(default) an assumption also escalates only. Underrelaxedan assumption may close the low-blast-radius routes —trusted-input,adversary-not-in-scope,unsupported-component,non-default-build, and a non-security-criticalproperty-disclaimed— as a provisional close: cite theQN, note it re-opens on challenge, and leave the §1.18 item open. - Security-critical floor (both policies). An assumption never licenses
KNOWN-NON-FINDING, aproperty-disclaimedwhose property issecurity-critical, ordependency-contract; those escalate unless documented / maintainer. - Silence floor (both policies, every provenance). A §1.12 disclaimer that rests on the absence of a statement rather than on