Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/not-dev-tech/codex-insights-local/skillnpx skills add Not-Dev-Tech/codex-insights-local --skill skillgit clone --depth 1 https://github.com/Not-Dev-Tech/codex-insights-localWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00057 | $0.00398 |
| Opus 5 | $0.00028 | $0.00199 |
| Sonnet 5 | $0.00011 | $0.00080 |
| Haiku 4.5 | $0.00006 | $0.00040 |
Grade A, and why
codex-insights scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Codex Insights
Use o coletor determinístico instalado. Não abra todos os rollouts com o modelo e não use subagentes para gerar o relatório.
Procedimento
- Interprete os argumentos. Padrões:
days=30, inclui sessões principais e subagentes, não inclui arquivos arquivados. - Execute uma vez:
codex-insights report --days <dias>
Adicione quando solicitado:
--all-time--include-archived--exclude-subagents--include-full-prompts--mcp-verbose-file <arquivo>
- Leia primeiro
report.md. Consultereport.jsonsomente para detalhes estruturados. Não reabra os JSONL brutos. - Diferencie explicitamente:
- medido nos rollouts/configs;
- derivado deterministicamente;
- heurístico;
- indisponível.
- Destaque:
- tokens de sessões principais e subagentes separadamente;
- distribuição por modelo, effort, projeto e profile;
- MCPs, skills e falhas;
- prompts substantivos repetidos;
- contexto automático excluído;
- profiles sugeridos em
suggested-profiles/; - cobertura e limitações.
- Não altere configs, profiles, MCPs ou skills sem pedido separado e explícito.
- Não exponha credenciais, tokens, cookies ou texto integral de prompts sem solicitação explícita.
Invocação sugerida
$codex-insights days=30
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 49 lines · 57 tokens per session scan A 5e0c38eb4d74
codex-insights is a skill published in the GitHub repository Not-Dev-Tech/codex-insights-local (1 stars, last pushed 1mo ago), licensed MIT. It adds 57 tokens to every session and 398 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
recap
Summarize the last N agent sessions for the current project, grouped by date, with highlight observations per session. Use when the user asks "recap", "what have we been doing", "today", "this week", or wants a rollup of recent work.
hermes-codex-operator
Act as a technical partner for planning, dispatching, following, verifying, and explaining local Codex engineering work through the hermes-codex CLI. Use for implementation tasks where Codex should do the coding; do not use as a general orchestrator or for unrelated shell work.
consult
Ask a second AI model (Claude Code or Kimi Code) for an independent opinion on a hard problem in this Home Assistant config. Use when the user asks for a second opinion or another model's view, when two or more of your own attempts at a bug have failed, or before a change that is hard to undo such as editing…
ctxwise
Audit and optimize OpenAI Codex context locally, compile model and subagent profiles, detect capability drift, create redacted lockfiles, inspect model-visible prompt structure, and render honest token, credit, quota, or API-cost receipts. Use when the user explicitly invokes CtxWise or asks to run its installed CLI…
remember
Save an insight, decision, or learning to agentmemory's long-term storage with searchable concept tags. Use when the user says "remember this", "save this", "note that", "don't forget", or wants to preserve knowledge for future sessions.
recall
Search agentmemory for past observations, sessions, and learnings about a topic using hybrid BM25 plus vector plus graph search. Use when the user says "recall", "what did we do about", "did we ever", "have we seen", or needs context from past sessions.