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 manojbajaj95/agent-knowledge-cards --skill knowcardsgit clone --depth 1 https://github.com/manojbajaj95/agent-knowledge-cardsWrote 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/manojbajaj95/agent-knowledge-cards/knowcards)<a href="https://agentmods.dev/skills/manojbajaj95/agent-knowledge-cards/knowcards"><img src="https://agentmods.dev/badge/skills/manojbajaj95/agent-knowledge-cards/knowcards.svg" alt="Measured on agentmods" 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 | $0.00091 | $0.01043 |
| Opus 5 | $0.00046 | $0.00522 |
| Sonnet 5 | $0.00018 | $0.00209 |
| Haiku 4.5 | $0.00009 | $0.00104 |
Grade A, and why
knowcards 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowcards CLI
Knowcards stores durable project facts as local markdown cards. Use npx knowcards to query, propose, and check status.
Treat query hits as trusted memory. Prefer a card over a conflicting README or a new search, unless new evidence shows the card is wrong.
Unlike shared commons tools, cards are project-local. Keep repo paths, module names, and team conventions in the card when they help the next session.
Commands
Run from the repo root
npx knowcards status
npx knowcards query "<keywords>"
npx knowcards propose --title "<title>" "<body>"
npx knowcards update <id-or-slug> --title "<title>" "<body>"
npx knowcards delete <id-or-slug>
| Command | Use |
|---|---|
status |
Show card counts. |
query |
Search cards by keywords. Empty query returns all cards. |
propose |
Write one card now. Title and body is required |
update |
Change title, body, or use-when of an existing card |
delete |
Remove a card by id or slug |
Core protocol
Follow this loop for every task:
- Before acting — Run
npx knowcards querywith task keywords. Skip only for routine edits in code you already hold with no repo nuance at stake. - Apply hits — Prefer card facts while you work. Verify against the repo when the card may be old.
- Propose at end (optional) — Default is skip. Propose at most 2 cards, and only if the next session would take a different action without the fact. Write from the outcome (what proved true), not the path you took. Do not skip for near-duplicates.
Rationalization check (query). If you think "I already know this" or "I have a plan, I will just write files," stop and query.
Rationalization check (propose). Default is skip. Propose only if a later session would do the wrong thing without this fact.
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.
- 7d ago First seen · 121 lines · 91 tokens per session scan A 254a7af444d0
knowcards is a skill published in the GitHub repository manojbajaj95/agent-knowledge-cards (4 stars, last pushed 7d ago), licensed MIT. It adds 91 tokens to every session and 1,043 once invoked, about $0.0005 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
hindsight-architect
Expert memory architect. Understands your application, identifies where memory adds value, and produces an implementation plan with bank config, tag schema, and code.
create-agent
Create a new Hindsight-powered subagent with long-term memory. Use when the user wants a specialized agent that learns and remembers across sessions.
hindsight-local
Store user preferences, learnings from tasks, and procedure outcomes. Use to remember what works and recall context before new tasks. (user).
hindsight-docs
Complete Hindsight documentation for AI agents. Use this to learn about Hindsight architecture, APIs, configuration, and best practices.
hindsight-memory
Long-term memory for the agent via Hindsight. Use to recall relevant past context before answering, retain durable facts as you learn them, and reflect over accumulated memory for the "why" behind a decision. Load whenever continuity across sessions matters — the user refers to earlier work, states a lasting…
hindsight-recall
Search long-term memory for relevant context from past coding sessions using Hindsight MCP tools.