362 packages matching “metacognitive-poisoning”
@rebase-lat/spectorio
v0.1.6 · 12 days ago
Trap-aware spec-driven development workflow (OpenSpec fork) as an installable agent skill
No known vulnerabilities
@vibo-dev/memory-shield
v0.1.5 · 28 days ago
Protect agent memory: snapshot before compaction, poison-scan for injected instructions, audit what changed. Zero dependencies, works with any memory backend.
No known vulnerabilities
pi-multi-account
v1.23.2 · 6 hours ago
Automatic multi-account failover & rotation for Pi Agent across Anthropic (Claude), OpenAI/ChatGPT Codex, Kimi For Coding, Cursor, Qwen/Alibaba, and Ollama. Auto-discovers authenticated accounts, grows the rotation on login, and drops accounts on logout,
No known vulnerabilities
dsh-daoing-memory
v0.1.20 · 1 month ago
Self-evolving memory for DeepSeek Harness (DSH): earned experiences, diary/fact semantic memory, concern tracking, and an append-only audit ledger.
No known vulnerabilities
@marketnow/sentinel-rules
v1.1.2 · 3 days ago
29 security rules for MCP servers & agents — tool poisoning, prompt injection, exfiltration, attack chains, stale-trust caching. Ships semgrep ruleset + zero-dependency lite scanner.
No known vulnerabilities
marinaedutlan
v0.8.2 · 3 months ago
Marina — Unified Metacognitive System. Professional CLI + lightweight unified core (DMA + Meta-DNA + MCP). `npm install -g marinaedutlan` gives you `marinaedutlan run`, `marinaedutlan doctor`, and `from 'marinaedutlan/core'`. Full visual lab available fro
No known vulnerabilities
sanitize-against
v1.8.0 · 23 days ago
Highly customizable, but simple to use, sanitization of objects and primitives.
No known vulnerabilities
ssvc
v0.4.1 · 1 year ago
TypeScript implementation of SSVC (Stakeholder-Specific Vulnerability Categorization). A prioritization framework to triage CVE vulnerabilities as an alternative or compliment to CVSS
No known vulnerabilities
@dutjr1/mcpwatch
v0.8.0 · 27 days ago
Security scanner for MCP (Model Context Protocol) servers and configurations
No known vulnerabilities
protofree
v1.0.0 · 6 years ago
Disable the __proto__ property for better security
No known vulnerabilities
@florentin-one/mcp-metacognitive-monitoring
v0.4.13 · 1 month ago
MCP server for diagrammatic thinking and spatial representation
No known vulnerabilities
@reactive-agents/reactive-intelligence
v0.16.0 · 18 days ago
Reactive Intelligence layer for Reactive Agents — entropy sensing, adaptive control, and learning
No known vulnerabilities
structured-thinking
v1.0.2 · 1 year ago
A TypeScript implementation of Sequential Thinking Server using MCP
No known vulnerabilities
llm-prompt-guard
v2.2.1 · 4 days ago
Sub-millisecond Layer-1 prompt-injection defense for TypeScript: detection triage, encoding-bypass normalization, quarantine/spotlighting, canary and output scanning. Zero dependencies.
No known vulnerabilities
septum
v0.1.1 · 4 days ago
Deterministic Bounded-Context, Structural File Catalog & Anti-Poisoning Partition for AI Agents
No known vulnerabilities
@mystilleef/vibe-cli
v0.4.2 · 1 month ago
Metacognitive AI agent oversight — pure CLI with JSON I/O
No known vulnerabilities
@rotalabs/probe
v0.0.2 · 2 months ago
Sandbagging detection via metacognitive probes
No known vulnerabilities
deep-thinker
v3.0.1 · 4 months ago
Advanced cognitive thinking MCP with DAG-based thought graph, 10 reasoning strategies, node aliases, session persistence, structured responses, auto strategy selection, and 17 tools with help & conclude
No known vulnerabilities
mcp-agentgate
v0.67.61 · 1 month ago
Scan, lock, and gate your MCP servers — npm audit + lockfile + CI drift gate for the MCP era
No known vulnerabilities
ken-you-think
v3.0.0 · 1 year ago
A single-tool MCP that guides coding agents through implementation thinking - TypeScript edition
No known vulnerabilities