Persistent project memory and intelligence layer for Codex and AI coding agents.
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Explicitly invoke existing CMI activation (initialize/scan project intelligence and, for Codex, manage bounded AGENTS.md and .codex/config.toml integration). Use only when the user explicitly asks to activate CMI, set up CMI integration, or configure CMI for Codex/generic (including Vietnamese activation requests).…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Apply the CMI Agent Operating System contract for substantive work: orient with explicit goal, scope and authorization boundaries; observe before editing; capture typed evidence; diagnose before symptom fixes; prioritize explicitly; implement the smallest coherent change; verify the original journey or contract…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Provide a bounded, evidence-labeled CMI ambient task brief by invoking existing Ambient Intelligence (MCP getambienttaskbrief or the documented project-local ambient CLI). Use when beginning substantive repository work or when the user asks for an ambient brief, task context, or what CMI knows about a request.…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Orchestrate the evidence-driven CMI Change Intelligence lifecycle (BEFORE start, DURING observe, AFTER complete/progress) for real implementation, refactor, or fix work. Use when the user asks to track a change with CMI, record changed files, finish or pause a Change with verification, or keep partial work active…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Read and surface existing Closing Intelligence for an already-closed CMI work session (alerts, CLEAN only when authoritative). Use when the user asks what CMI flagged when work ended, blockers/reminders from a closed session, or to show Closing Intelligence / final CMI alerts (including Vietnamese closed-session…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Resume known unfinished project work from durable CMI session handoff evidence, current open findings, current active Changes, and a fresh repository baseline re-check. Use when the user asks to continue, resume, pick up previous work, làm tiếp, or identify what unfinished CMI-tracked work remains. Strictly read-only…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Preserve evidence integrity for CMI-assisted work by establishing a baseline, maintaining an evidence ledger, attaching addresses to important claims, separating observations from inference and recommendations, and reporting focused/repository/CI/live/release gaps truthfully. Thin adapter over existing CMI evidence…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Provide a concise read-only assessment of whether CMI project evidence is currently usable, stale, missing, degraded, or blocked before relying on it. Use when the user asks if CMI is healthy, whether intelligence/evidence is current, why CMI is blocked, if the graph/index is stale, or to check evidence health before…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Read-only review preparation for durable CMI memory—surface stale, review-required, untracked, inactive, or blocked entries so a human can decide next steps. Use when the user asks which memory needs review, what knowledge is stale, to audit project memory, or to inspect inactive/blocked memory (including Vietnamese…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Review or edit user-facing prose for clarity, specificity, repetition, unsupported attribution, generic AI-like writing patterns, and unnecessary formatting while preserving the author's meaning and CMI evidence/provenance semantics. Use as a final quality pass for documentation, reports, handoffs, release notes, and…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Prepare an evidence-bounded release-readiness assessment for an exact revision by checking scope, acceptance criteria, behavior, findings, packaging, repository gates, CI/live evidence, rollback and approval boundaries. Thin adapter over existing CMI Change, Session, findings, provenance and trust surfaces; it never…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Create or adapt portable Agent Skill artifacts from concrete, evidenced workflows while preserving CMI authority, evidence, lifecycle, and runtime boundaries. Use when a reusable workflow should become a SKILL.md package or an external/vendor-specific Skill needs portability refactoring. It does not install or…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Discover and assess external Agent Skill artifacts when the current runtime or CMI Skill set lacks a needed workflow. Use to find candidate SKILL.md packages from authorized sources and decide whether to reject, evaluate, or adapt them. Advisory only; it does not auto-install, auto-activate, execute, trust, or promote…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Evaluate whether an existing library, tool, project-local component, standard capability, or other reusable solution should be adopted before custom implementation. Use for well-trodden implementation problems where reuse could reduce cost or risk. Evidence-bounded advisory Skill only; it does not install…
lenhonbp/codex-memory-intelligence
Skill Claude CodeCodex
Orchestrate the durable CMI work-session lifecycle (start, observe meaningful progress, finalize) and surface authoritative Closing Intelligence/handoff. Use when the user asks to track work as a CMI session, record blockers/decisions/progress, or finalize/close the current session (including Vietnamese…