Security Audit Skills Package
Security-audit skill pack for DeepSeek Harness (dsh): 8 agent skills - secret scan, dependency audit, supply-chain review, prompt-injection review, audit orchestration, threat modeling, vuln intel, incident response.
Project overview
Security-audit methodology for DeepSeek Harness — eight agent skills, zero runtime code. secret scanning · dependency audit · supply-chain review · prompt-injection review · audit orchestration · threat modeling · vuln intel · incident response
Installed together with a tool-type security plugin, the two compose: the tool runs the scan, the skill drives interpretation, triage, and the report — the model follows this pack's methodology while calling the tool plugin's tools.
DSH's local skill provider scans four roots by rank — lower rank wins same-name conflicts within a layer: Rank · Root · Scope Rank: 100 · Root: /.dsh/skills · Scope: Project-scoped, travels with the repo Rank: 200 · Root: /.agents/skills · Scope: Project-scoped, shared agent directory Rank: 400 · Root: /skills ($DSHHOME or /.dsh) · Scope: User-scoped, DSH-only Rank: 500 · Root: /skills ($DSHAGENTSHOME or /.agents) · Scope: User-scoped, cross-agent