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 skills add WilliamJnsson/NarratorDB --skill remembergit clone --depth 1 https://github.com/WilliamJnsson/NarratorDBWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/williamjnsson/narratordb/remember)<a href="https://agentmods.dev/skills/williamjnsson/narratordb/remember"><img src="https://agentmods.dev/badge/skills/williamjnsson/narratordb/remember/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/williamjnsson/narratordb/remember"><img src="https://agentmods.dev/badge/skills/williamjnsson/narratordb/remember.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00039 | $0.00671 |
| Opus 5 | $0.00019 | $0.00336 |
| Sonnet 5 | $0.00008 | $0.00134 |
| Haiku 4.5 | $0.00004 | $0.00067 |
Grade A, and why
remember 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 8d ago.
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
Remember with NarratorDB
For a normal single-item request, use this exact interaction:
- Build one self-contained statement that will make sense without the current
transcript. Put attribution and category detail in
content. - Send one short progress update:
Saving that to NarratorDB… - Call
rememberexactly once with:content: the durable statement;scope:projectunless the user explicitly requests a cross-project preference, thenglobal;source:userfor a user-stated fact, preference, decision, convention, or explicit remember request.
- Interpret the structured result internally and finish with one polished line:
- stored:
Remembered for this project: <brief description>. - duplicate:
Already remembered for this project: <brief description>. - failure:
I couldn't save that to NarratorDB: <short actionable reason>.
- stored:
For scope="global", say Remembered globally or Already remembered globally instead of for this project.
source is a strict enum. Its only valid values are user, assistant,
system, and memory. Never pass prose such as explicit user-stated convention, a category name, or an attribution label as source. Use
assistant only for a durable outcome the assistant actually completed;
otherwise default to user for this skill.
Example for “Remember that this project uses pnpm”:
{"content":"For this project, the user requires pnpm for package operations.","scope":"project","source":"user"}
Do not call status or recall before a normal write. Do not retry a failed
call with a guessed schema. Do not print or restate raw JSON, message_id,
workspace_id, ingest_ms, or other tool fields unless the user asks for
diagnostics. Codex controls its native tool spinner; do not simulate an
animation with repeated messages.
Keep these content rules:
- Preserve whether the user stated something or the assistant completed it.
- Include the project or component when needed to prevent ambiguity.
- Split unrelated facts into separate memories only when the request genuinely contains multiple durable items; use one call per item.
- Never store passwords, API keys, access tokens, private keys, authentication cookies, or raw secret-bearing files. If the request contains a secret, store only a safe statement such as “Deployment credentials are configured externally.”
- Never claim success unless the tool reports
stored=trueorduplicate=true.
For an explicitly requested completed-session summary, make one
remember_session call instead. Message roles must also use only user,
assistant, system, or memory. Keep goals, decisions, changed components,
verification results, unresolved issues, and next steps; exclude filler and
speculation. Use the same one-update, one-call, one-line-result pattern.
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.
- 8d ago First seen · 63 lines · 39 tokens per session scan A 252dc21c8452
remember is a skill published in the GitHub repository WilliamJnsson/NarratorDB (0 stars, last pushed yesterday), licensed MIT. It adds 39 tokens to every session and 671 once invoked, about $0.0002 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
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
mem0-oss-to-platform
Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.
ha-data-stores
Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…
establishing-project-context
Use when the user asks to establish shared project language, or project work exposes a conflicting, renamed, or deprecated domain term that needs active semantic modeling. Routine small tasks stay on the fast path.