SEMANTIC DIRECTORY · plugin-directory-semantic-v3-19
Agent Runtime Behavior Guardrails
约束智能体执行过程和完成声明，包括固定语言、工作纪律、防提前宣称完成、死循环硬中断、自检规则、科研防自欺和运行时行为监控；不负责提示预设设计或Request传输策略。
Leaf membership
17 plugins
autonomous-circular-agentjonah791/autonomous-circular-agentCatalog Analyzed自主循环智能体最小可行方案 - Autonomous Circular Agent MVP: loop = existence itself (verified on DeepSeek Harness, 2026-08-15)CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals1 starsbreakthrough-harnessGuoCheng24/breakthrough-harnessCatalog AnalyzedMake your research agent hard to fool — the discipline layer for agentic research: breakthrough loop, hard-to-fool harness checklist, claim-polarity gates, engineering rules with their tuitionCompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals1 starsconceptnet-dsh-pluginwushu75/conceptnet-dsh-pluginCatalog AnalyzedConceptNet intent classification plugin for DeepSeek Harness. Classifies enterprise voice and text commands into 4 execution layers — Basic, Context-Aware, Predictive, Autonomous — at 100% accuracy. 9 languages. Token-free. No LLM required.CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals0 starsdonechecklinkage18/donecheckCatalog AnalyzedA plugin for DSH that checks whether your work has actually been completed. It automatically loops over the model’s responses to check for errors or shortcomings.CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals1 starsdsh-advisorslhssb/dsh-advisorCatalog AnalyzedIndependent-model advisory review for DeepSeek Harness: after each tool step, a reviewer model audits the agent's operations and injects concerns/guidance into the next step.CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals1 starsdsh-advisor-pluginleeyoung1/dsh-advisor-pluginCatalog AnalyzedPairs the running executor with a stronger reviewer model through zero-parameter advisor() calls and step-based patrol checks, returning plan, correction, or stop guidance.CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signalsNo public usage signalsdsh-agent-loop-guard@goodandready/dsh-agent-loop-guardCatalog AnalyzedFail-closed runtime tool-call loop guard for DeepSeek Harness.CompatibilityDSH 0.1.6-alpha.2 · webDependency audit0 reported in bounded auditArtifact0.2.14Evidence updated2026-09-24Public signals0 stars · 980 downloadsdsh-anchored-monitor@a9i5k4/dsh-anchored-monitorCatalog AnalyzedA whip for DeepSeek V4 Pro: it watches the reasoning fingerprint of every thinking block and pulls the model back when it slips from focused "We will / I will" mode into scattered "let me" mode.CompatibilityDSH 0.1.6-alpha.2 · webDependency audit0 reported in bounded auditArtifact0.3.1Evidence updated2026-09-19Public signals10 stars · 1151 downloadsdsh-j-spaceHarzva/dsh-j-spaceCatalog AnalyzedDSH plugin from Harzva/dsh-j-spaceCompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals1 starsdsh-loop-continuedsh-loop-continueCatalog AnalyzedResumes an agent turn that ended after narrating its next action without calling a tool: the guard replays the turn log on agent/turn-stopping, asks a judge model one true/false question, and steers the same turn to run one more step.CompatibilityDSH 0.1.6-alpha.2 · webDependency audit0 reported in bounded auditArtifact0.2.0Evidence updated2026-09-21Public signals2 stars · 648 downloadsdsh-loop-detectorviego-qiin/dsh-loop-detectorCatalog AnalyzedDead-loop / repetition detector for DeepSeek Harness agents — hard-interrupts text loops and same-topic search retries that soft reminders miss.CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals2 starsdsh-plugin-j-spacekolawong/dsh-plugin-j-spaceCatalog AnalyzedJ-Space Cognition Suite V3.6 - Inference-time cognitive control and deep reasoning plugin for DeepSeek Harness (DSH)CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals6 starsdsh-thinking-loop-guardunknowbug/dsh-thinking-loop-guardCatalog AnalyzedDetect & break thinking-chain loops in DSH agents at the turn boundary (no proxy). Ported from ollama-loop-guard.CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals2 starsdsh-verdict-guardsagetta1/dsh-verdict-guardCatalog AnalyzedNative DeepSeek Harness plugin: a turn that states a verdict without evidence is held back and steered for proof.CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals0 starsJ-Space-Cognition-Suite-VNext-Governororangeofcarl0-sys/J-Space-Cognition-Suite-VNext-GovernorCatalog AnalyzedJ-Space Cognition Suite vNext Epistemic Governor — 独立Plugin原型,基于 J-Space Cognition Suite V3.6 的行为设计衍生,分离执行信心与认知信心。CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals1 starsPSA-coreSiliconPsycheLabs/PSA-coreCatalog AnalyzedPSA is a multi-classifier behavioral analysis engine for LLM responses. It classifies every sentence into behavioral postures, then derives metrics from the posture sequences to detect adversarial stress, sycophancy, hallucination risk, persuasion techniques, and input pressure — in real time.CompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals4 starsWAM-Frameworkzbbsdsb/WAM-FrameworkCatalog AnalyzedWait a minuteCompatibilityDSH 0.1.6-alpha.2 · webDependency auditNot testedArtifactUnresolvedEvidence updated2026-09-27Public signals2 stars
Categories are public semantic index facts, not a final recommendation. Compatibility, permissions, and source evidence remain separate.