By SpeakToJade
Claude-to-Claude messaging protocol for multi-agent coordination and communication within Claude Code environments.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub joshuarweaver/cascade-ai-ml-agents-misc-1 --plugin speaktojade-nclaudeThe bridge between Claude's working memory and Basic Memory's durable knowledge graph — session briefings, pre-compaction checkpoints, and capture reflexes
Foundation plugin: analyst trio personas, cross-cutting skills, canonical Shared_Utils source-of-truth, subagent-stop hook
Use fewer output tokens with short steps, quick insights, and clear next actions.
AlgoKiller — ARM64 trace evidence analysis & cipher algorithm recovery. Bundles the AlgoKiller methodology as skills + a local MCP server (28 tools) that drives the native ak_search engine over GB-scale trace files. v1.2.0 adds two trace tools targeted at OLLVM-flattened binaries — `trace_function` (PC-level invocation analyzer with depth-counter ret boundary, replaces manual trace_search + trace_context loop for stripped helpers) and `trace_immseq` (anchor-driven prev_reg sequence extraction, recovers table-consumption order from jump-table-dispatcher functions). Formalizes R9 (`regN=X -> regN=Y` prev/new semantics — single most common LLM misread in obfuscated code) and FIX #8 (algorithm-name claims require trace_cryptoinstr/trace_constscan/trace_immseq/trace_function fingerprint corroboration before conclude(>=medium)). Loosens five over-strict gates: trace_search limit 100→500, B2 same-args rule exempts hint-driven fallback paths, cross-session [H<n>] references become non-blocking warnings instead of hard errors. Tests 128 → 149 PASS.