dsh-decision-effect-proof
Run the following command in DeepSeek Harness:
dsh plugin install dongsheng123132/dsh-decision-effect-proof
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install dongsheng123132/dsh-decision-effect-proof in DeepSeek Harness to install this plugin; source is available at https://github.com/dongsheng123132/dsh-decision-effect-proof
About this plugin
Across a multi-layer security stack, approval engines, policy evaluators, and action runtimes each guard their own boundary, yet a narrow gap remains that none of them can address on their own: authorized does not mean executed. dsh-decision-effect-proof exists to close that gap. It performs an offline, deterministic reconciliation between a recorded authorization decision and its recorded effect envelope, surfacing denied actions that carry effects, authorized actions that never settled, request-state-policy mismatches, missing confirmations, replayed idempotency digests, duplicates, and out-of-order evidence, each flagged by a stable finding code so audits remain reproducible.
By design the plugin does not cross into adjacent territory. It is not an approval engine, policy evaluator, action runtime, signature format, or prompt wrapper. It consumes only explicit body-free evidence, namely SHA-256 digests and bounded identifiers, and never touches request arguments, tool output, business text, prompts, or secrets. Reports are content-addressed and verified by read-back after publication, guaranteeing that what you archive is exactly what was produced.
Teams that have already layered DSH approval and permission-rule plugins and now need a tamper-evident, offline consistency proof for compliance will find a natural slot for this tool. Security analysts doing incident retrospectives can also use it to determine, in seconds, whether a given authorization trajectory shows an execution break, rather than relying on guesswork.
Use Cases
- Off-line audit of whether a recorded authorization was actually executed
- Incident retrospectives to pinpoint execution breaks in an authorization trail
- Periodic integrity checks on the sequencing and consistency of approval evidence
Best For
- SecOps and compliance audit teams
- Platform teams running multi-layer DSH approval and permission stacks
- Security architects in finance or healthcare requiring an offline evidence chain
Related Plugins
Adds an Auto approval mode on the official workspace-write sandbox, classifying semantic risks, asking when ambiguous, and denying destructive operations.
Undo/rollback system for DSH: config & plugin snapshots, one-click undo/redo/restore, message-level revert, secret masking, safe mode, and offline WebUI/GUI/CLI rescue tools for when DSH won't start.
Packages all 87 SKILL.md files from upstream reverse-skill as a DeepSeek Harness plugin that auto-registers them for authorized reverse engineering, penetration testing, and security research.
Dockyard DSH is a native DeepSeek Harness plugin that unifies official OAuth/client sessions for Codex, Antigravity, Grok, Claude, and Cursor, providing account pool, model catalog, and quota status.