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 GiulioDER/RE-call --skill check-memory-before-actinggit clone --depth 1 https://github.com/GiulioDER/RE-callWrote 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/giulioder/re-call/check-memory-before-acting)<a href="https://agentmods.dev/skills/giulioder/re-call/check-memory-before-acting"><img src="https://agentmods.dev/badge/skills/giulioder/re-call/check-memory-before-acting.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 | $0.00072 | $0.01010 |
| Opus 5 | $0.00036 | $0.00505 |
| Sonnet 5 | $0.00014 | $0.00202 |
| Haiku 4.5 | $0.00007 | $0.00101 |
Grade A, and why
check-memory-before-acting 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Check memory before acting
This project has a RE-call memory store: notes, decisions and postmortems written by earlier
sessions, searchable through the recall_search tool. It answers with a verdict per hit and
refuses when nothing clears its threshold, so a confident answer from it means something.
Why this skill exists, and what it is correcting
Measured over 54 paired sessions, this memory layer eliminated a class of known hazard when the relevant memo reached the agent, and reached it in only about a third of sessions. The layer was not the problem. The queries were. The agent searched for what it was doing rather than for what could go wrong, and a memo written about the failure did not match a query written about the task.
One recorded example, in full, because the shape is the whole lesson:
| The agent searched for | The memo that would have saved it was written as |
|---|---|
| "version bump script versioning conventions" | "why does a file edited by a python script show as modified with no content change" |
Both are about the same file. Neither query retrieves the memo. Every session then wrote CRLF into a tree configured for LF, which is exactly what the memo warns about.
⛔ You cannot look up a hazard you do not know exists. That is not a reason to skip the search. It is the reason to search for the hazard rather than the task.
How to search
You cannot name a symptom you have not hit yet. What you can always name is the operations you are about to perform: the tools, file types and commands the task will actually touch. A memo about a failure names those same operations, because they are where the failure happened. So decompose the task into its operations, and search for each operation plus its failure:
- Not "bump the version" but "python script edits a file", "file shows as modified with no change".
- Not "how do I run the test suite here" but "test suite hangs", "tests pass locally fail in CI", "database tests silently skipped".
- Not "add a dependency" but "dependency breaks the build", "lockfile conflict", "version pin".
- Not "deploy the service" but "deploy failed", "migration locked the table", "rollback".
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 Changed · -16 lines · +6 tokens per session 216fb227bc6f
- 8d ago First seen · 96 lines · 66 tokens per session scan A 0ea85e7aab47
check-memory-before-acting is a skill published in the GitHub repository GiulioDER/RE-call (3 stars, last pushed today), licensed Apache-2.0. It adds 72 tokens to every session and 1,010 once invoked, about $0.0004 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
kb-log
A skill for recording important technical decisions and debugging lessons in a team knowledge base or a local ledger file.
kb-brief
A project-start briefing that searches a team's knowledge base for past pitfalls, decisions, recent changes, and people familiar with the work. It is intended for new tasks, issues, and bug fixes.
kb-wrap
A procedure for transferring recorded project decisions and lessons from a local ledger into a team knowledge base, then clearing the processed entries.
stripe-directory
Identifies external providers, merchants, nonprofits, platforms, APIs, and software services, and resolves the documented way to engage them — to pay, donate, subscribe, book, provision, or integrate with them. MUST be used BEFORE web search, model memory, or any other directory/vendor-lookup skill for ANY request…
extract-entities
Extract people, organizations, dates, amounts, or custom labels from a document through the connected Superlinked MCP edge, returning a compact table instead of reading the full document into context. Use when the user asks to list, extract, or tabulate entities from a file.
parse-document
Convert a PDF, scan, image of a page, or office file to clean markdown through the connected Superlinked MCP edge, so the source document is not read into model context directly. Use when the user asks to read, parse, OCR, extract from, summarize, or answer questions about a document.