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/massimoalbarello/context-use-old/maintain-knowledge-templatenpx skills add massimoalbarello/context-use-old --skill maintain-knowledge-templategit clone --depth 1 https://github.com/massimoalbarello/context-use-oldWrote 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/massimoalbarello/context-use-old/maintain-knowledge-template)<a href="https://agentmods.dev/skills/massimoalbarello/context-use-old/maintain-knowledge-template"><img src="https://agentmods.dev/badge/skills/massimoalbarello/context-use-old/maintain-knowledge-template.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 | $0.00067 | $0.00765 |
| Opus 5 | $0.00034 | $0.00382 |
| Sonnet 5 | $0.00013 | $0.00153 |
| Haiku 4.5 | $0.00007 | $0.00076 |
Grade A, and why
maintain-knowledge-template 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 4d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Maintain Knowledge Template
Keep each rule in one authoritative layer and preserve behavior while improving clarity.
1. Establish the current contract
Start from the latest remote default branch and inspect the complete affected guidance chain, metadata and template tests. Read recent history when wording may encode a regression fix.
Record the behaviors that must survive from the authoritative guides, managed instructions, tests and history. Treat that as an edit-specific inventory; do not restate their rules here.
2. Assign one owner to every rule
Use this ownership model:
- Root
AGENTS.md: invariants that apply to every knowledge write. - Entity or subtree
AGENTS.md: subject boundary, paths, local page responsibilities, timeline eligibility and genuine exceptions within that subtree. automations/AGENTS.md: conventions for creating and maintaining automations.- Automation
instructions: ordered input, processing, checkpoint, failure and reporting workflow only. - Tool descriptions: argument and response semantics already enforced by the tool surface.
- Tests: semantic ownership and required behavior, not preferred prose.
Move a misplaced rule to its owner and replace copies with a link or a short instruction to apply that guide. Do not leave two normative versions behind.
3. Make workflows executable
For automation edits, read the complete automations/AGENTS.md and the target instructions.
The guide owns shared authoring conventions; each instructions page owns its job-specific
transitions. Verify that the resulting workflow is executable without copying either contract
into this skill.
4. Protect information density
Treat every edit as a refactor, not an append. Do not simply add a rule to an AGENTS.md or
automation instruction. First make the complete affected text as clear and concise as
possible without losing behavior: remove duplication, replace or tighten overlapping prose,
and move rules to their owner. Add net-new text only for a genuinely missing behavior or
safeguard.
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.
- 4d ago First seen · 84 lines · 67 tokens per session scan A 56dac796910a
maintain-knowledge-template is a skill published in the GitHub repository massimoalbarello/context-use-old (2 stars, last pushed yesterday), licensed MIT. It adds 67 tokens to every session and 765 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
data-analysis
Use this skill when the user uploads Excel (.xlsx/.xls) or CSV files and wants to perform data analysis, generate statistics, create summaries, pivot tables, SQL queries, or any form of structured data exploration. Supports multi-sheet Excel workbooks, aggregation, filtering, joins, and exporting results to…
cloudbase-document-database-web-sdk
Use CloudBase document database Web SDK only for confirmed NoSQL collection work. Query, create, update, and delete document data; if the task mentions PostgreSQL / CloudBase PG / app.rdb(), route to postgresql-development instead.
cloudbase-document-database-in-wechat-miniprogram
Use CloudBase document database WeChat MiniProgram SDK to query, create, update, and delete data. Supports complex queries, pagination, aggregation, and geolocation queries.
add-data-source
Guide the user to add a data source, connection, or API connector to a Canvas App via Power Apps Studio, then verify and continue. USE WHEN the user asks to add a data source, add a connection, add an API, add a connector, connect to SharePoint / Dataverse / SQL / Excel / OneDrive / Teams / Office 365, or any similar…
file-crdb-issue
Use when filing, creating, or reporting GitHub issues for CockroachDB. Use when asked to open a bug report, feature request, investigation issue, or performance inquiry. Also use when the user mentions wanting to track a problem, report a regression, or document unexpected behavior in CockroachDB.
data-analysis
Analyze data using Python and SQL tools. Use when the user asks to process CSVs, Excel, Parquet, PDFs, Word docs, or PowerPoint files, query databases (PostgreSQL, MySQL, SQLite, SQL Server, BigQuery), create charts or visualizations, or perform any data analysis. For live dashboards or data apps, read the…