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/coldzero94/cold-frame/remember-factsnpx skills add coldzero94/cold-frame --skill remember-factsgit clone --depth 1 https://github.com/coldzero94/cold-frameWhat 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 | $0.00058 | $0.00348 |
| Opus 5 | $0.00029 | $0.00174 |
| Sonnet 5 | $0.00012 | $0.00070 |
| Haiku 4.5 | $0.00006 | $0.00035 |
Grade A, and why
remember-facts 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 2d 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 durable facts to Coldframe
This workspace has persistent, local-first memory via Coldframe (the add_memory tool and automatic
recall). Recall is automatic — memories are injected at session start and per prompt, so you don't
need to search unless the user explicitly asks what you remember.
Your job is capture, and it costs nothing extra: doing it inside this conversation rides the interactive session, not a separate metered budget. So:
- When the user states a durable fact — a preference, decision, convention, identity/contact
detail, or a correction of an earlier fact — call
add_memorywith one concise, self-contained fact (one call per fact). Capture silently and keep going; don't announce it or ask permission. - Do not capture: transient task requests, questions, chit-chat, or anything that looks like a secret/credential (Coldframe blocks secrets, but don't send them).
- Phrase each fact as a standalone statement (e.g. "The user deploys this repo with ship.sh", not "ok I'll use that"). Coldframe de-duplicates, so restating a known fact is harmless.
A keyless backstop also captures durable facts from the transcript, so anything you miss is still caught — but your in-context judgment is higher quality, so capture what you notice.
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.
- 2d ago First seen · 25 lines · 58 tokens per session scan A bb2297fe650d
remember-facts is a skill published in the GitHub repository coldzero94/cold-frame (0 stars, last pushed 14d ago), licensed Apache-2.0. It adds 58 tokens to every session and 348 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
hr-onboarding
A new-hire onboarding plan as a single page — first week schedule, buddy + manager intro, learning track, equipment checklist, and "you're set when…" outcomes. Use when the brief mentions "onboarding", "new hire", "first week plan", or "入职".
html-ppt-taste-brutalist
16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).
new-changelog
Create the next desktop changelog entry when asked to add a changelog file or prepare the next release note under packages/changelog/content. Use this when the task is specifically about determining the next version and creating the markdown entry.
self-awareness
Inspect Open Science's JavaScript control REPL, discover managed Project files, Sessions, and Agent Frames, and safely feature-gate host. calls with host.capabilities(). Use when an Agent needs to discover available host APIs, locate an Artifact or Upload Version, diagnose a Session, or read a Frame transcript in the…
ligandmpnn
Inverse-fold a backbone with ligand, nucleic-acid, and metal context using LigandMPNN (Dauparas et al. 2023, github.com/dauparas/LigandMPNN). Reach for this skill to redesign the residues lining a binding pocket around a bound small molecule or cofactor, to design metal-coordinating sites where the geometry must be…
publish
Publish the Octop Python package: cut a release branch from develop, bump version, update CHANGELOG, open a PR to main; after merge, Actions tag on main (PyPI / Docker Hub) and sync main into develop. Use when the user asks to publish, release, bump version, cut a release, or run /publish.