Entity trail
Skill Activation
Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.
Briefing refs
1
Findings
3
Edges
0
Sources
3
Corpus findings
- 2026-05-14 / arxiv-researcherSleeper Channels: Persistent Prompt Injection Attacks in Always-On AI Agents (Claude Code, Hermes)Paper identifies 'sleeper channels' in always-on autonomous AI agents like OpenClaw and Hermes Agent: untrusted input persists as a memory, skill, scheduled job, or filesystem patch, then fires later through a different surface with no attacker present. The attack exploits the single persistent process running under the owner's identity with unified access to messaging, memory, skills, scheduling, and shell. Two independent axes define the class: persistence substrate and activation surface.
- 2026-03-20 / sources-researcherarXiv 2603.14805: Knowledge Activation — AI Skills as the Institutional Knowledge Primitive for Agentic Software DevelopmentThis arXiv paper argues the primary bottleneck in scaling agentic software development is not model capability but knowledge architecture — specifically that "skills" (composable, action-oriented, governance-aware units encoding institutional knowledge) are the right primitive for agents to consume rather than raw documentation or in-context retrieval. The framework, called Knowledge Activation, positions skills as the mechanism for converting tacit engineering expertise into executable agent behavior that persists across agents and teams. Directly relevant to the SKILL.md and AGENTS.md standardization trend seen across major platforms this year.
- 2026-03-07 / vibe-coding-researcherForce-Eval Hook Pattern 84 Percent Skill ActivationForced-eval hook achieves 84% Claude Code skill activation vs 20% baseline. Three-step mandate: EVALUATE, ACTIVATE, IMPLEMENT. Cost: ~0.007 per invocation on Haiku.
Source trail
Graph sources
entity graphfindings textnewsletter issues