Unverified50% confidenceFactExact time
Armor has three output_mode values: text (the default for Chat Completions and Responses, marked with <openrouter_armor_blocked>), custom_item (Responses, producing an openrouter:armor_blocked output item), and claude_code_compatibility (Messages, preserving tool_use and marking flagged in safeguard_results).
1
Sources
50%
Confidence
Long-term
Relevance
10/19/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedClearly separating task instructions from pasted material helps guard against prompt injection attacks.58% similarUnverifiedRunning the cslcore venom command scans the code repository in read-only mode, identifying @tool-decorated functions, StructuredTools, LangGraph nodes, as well as OpenAI/Anthropic tool schemas, MCP servers, and webhooks.57% similarUnverifiedCodeSmith makes example-driven generation a built-in mechanism; in crates/agent-runtime/src/skills/system.rs it embeds example skills such as SKILL_CREATOR_BODY, DELEGATE_BODY, V4_BEST_PRACTICES_BODY, PLUGIN_CREATOR_BODY, and MCP_BUILDER_BODY via include_str!, and on first startup installs them automatically and performs version management and incremental updates through BUNDLED_SKILL_VERSION.57% similarUnverifiedThe edges in BlastShield's graph model include HAS_SESSION, CAN_ACCESS, RUNS_AS, CAN_ASSUME, and CAN_READ, and the edges carry attributes such as route identifiers, containment and response status, permission assessments, and sensitivity.57% similarUnverifiedIn OpenCode V2, ~/.claude/skills is read automatically, but skills inside plugins are not loaded automatically, and the skills directory in the plugin cache must be manually written into the configuration; when IDs conflict the latter overrides the former, so personal skills should be ordered after plugins.56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1065092API
curl https://kongchang.com/api/v1/knowledge/claims/1065092MCP
get_claim(id=1065092)