agents/_lanes/555-ASI/skills/asi_memory_validate/SKILL.md
--- name: asi_memory_validate agent: 555-ASI namespace: asi_* cluster: MEMORY trigger: "When 555 is about to use a recalled memory as a premise for a position, a synthesis, or a seal recommendation, and the memory's currency must be confirmed against current evidence" capability: "Returns a validation verdict (current / stale / contradicted / expired) for a recalled memory, with the specific drift detected and a recommended disposition (use as-is / refresh / deprecate)" mcp_tools_underneath: "fo
npx skillsauth add ariffazil/openclaw-workspace asi_memory_validateInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Memory without validation is just narrative. This skill is the one that asks, of any recalled memory: is this still true? It does not assume that a sealed memory is a current memory. Sealed means "this is the artifact as of that moment"; current means "this still reflects the state of the world." The two are different, and confusing them is one of the most common federation failure modes. 555 runs this skill every time it is about to feed a memory into a position, to prevent stale-claim errors.
The validation is three-axial. Temporal currency — has the relevant window passed? A scar from 2025 may still be structurally current; a vitality reading from 30 minutes ago may be stale if the operator's state has shifted. Evidence currency — has the underlying evidence been updated, revised, or contradicted by later events? Sovereign currency — has the operator or a peer organ explicitly retracted, amended, or superseded the memory? Each axis is checked, and the worst axis determines the verdict. The skill never "averages" currency — a single axis returning contradicted collapses the memory to contradicted regardless of how strong the other axes are.
The skill is also the gate that prevents memory laundering — a failure mode where a 555 draft cites a memory that supports its position while ignoring memories that contradict it. The validation report names what was checked, what was found, and what was NOT found (the absences matter as much as the presences, especially for a sovereign-only signal that has since been retracted — the skill is biased toward remembering what the operator wanted forgotten, in the way they wanted it forgotten, not toward letting a peer organ "un-remember" it by ignoring the record).
Failure modes: (a) the recalled memory has no chain head — validation is blocked, and the memory is treated as unverifiable; (b) the current evidence is itself low-grade (bronze) — validation returns indeterminate rather than picking a side; (c) the operator has asked that this memory never be re-validated against current evidence (a "frozen" memory) — the skill honours the freeze and returns the memory as frozen_current, with the freeze flag preserved in the local ledger. The skill never refreshes a memory silently; a refresh is a constitutional event that must be sealed.
testing
OpenClaw edge agent bridge — operational triage, doctor, restart, and A2A bridge routing for the federation edge (Telegram surface). USE WHEN: "openclaw unhealthy", "gateway down", "edge bot not responding", "a2a bridge disconnected", "watchdog tripped", "openclaw doctor", "openclaw restart". NOT for token/security audit — use FORGE-telegram-audit.
tools
Generate images, videos, TTS, voice clone, and music via MiniMax MCP server. Use when user asks to "draw", "generate image", "create picture", "make a photo", "text to image", "image generation".
testing
Single load-bearing constitutional-judgment skill. Routes all F1–F13, verdict, hold, seal, scope, authority and floor-check calls through the live arif_judge surface. Replaces 7 overlapping predecessors (arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge).
development
MANDATORY LSP grounding gate BEFORE any code mutation on .ts, .py, .js, .tsx, .jsx files. Forces the agent to read real-time compiler diagnostics and structural project context before editing — eliminating blind guesses and anchoring every mutation in F2 (TRUTH). Routes through arifOS kernel (:8088) for centralized gate logic.