repomagpie@0.2.1
VERIFIED· install onlyPublisher claims: "A local MCP server that lets coding agents recall your Markdown notes and verdicts on packages before installing them, and search your personal and shared project journals by concern in any language. It also surfaces alternatives and flagged dependencies so agents ask before repeating an install you noted to avoid." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Nothing executable in the bundle ran; the trace covers the install step only.- Tested
- every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
What this means for the label: Verified here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.
Before you give it your token
- It will ask for
- no credential variables read in the code
- Where it talks to
- no vendor host named in the codeAlso crates.io — not the publisher's domain; check what for before connecting.
- What the honeypot saw
- no credential files read · no network
- What we could not test
- use of the skill by an agent
Connect it under the guard. Nothing executable in the bundle ran; the trace covers the install step only. The code names no credential variable; the guard pins the network to what the honeypot saw and logs every call.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (1 scenario; nothing executable in the bundle ran; the trace covers the install step only) on 2026-11-08 and scored it 80/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 1 scenarios · 0 deviationsStatic findings
<summary><strong>Where are my notes, and can I edit them by hand?</strong></summary> Each note is one Markdown file with YAML frontmatter, in `~/.magpie/notes/`Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.Suggest in repomagpie ui: microsoft/playwright first, matched on the @playwright/test dependency; pdfkit listed last as in use with an avoid note, with pdf-lib Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.How RepoMagpie works: your agent is about to run npm install pdfkit; the skill or the MCP tool recall checks first, and the hook checks the install command as aFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><strong>Does RepoMagpie send my notes anywhere?</strong></summary> No. Your journals are Markdown files on your machine, and there is no telemetry. TheFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><strong>Does the hook block installs?</strong></summary> Never for good, and it never approves one. When a note says to avoid the package, the hook askFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><strong>Does RepoMagpie use AI?</strong></summary> Not a model of its own: its commands match names and words, and call no model. The AI is the coding Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><strong>How do I remove it?</strong></summary> Run `repomagpie setup --remove`. It takes out only RepoMagpie's own MCP entries, hooks and skill foldersFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.if (tableName(h) === `mcp_servers.${name}.env`) continue;Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.// and Claude Code's, then the agent folders npx skills or a person may have used.Fix: Pin versions and sources; vendor anything fetched at install time.// npx cache. Node runs with code generation from strings disallowed (decision 0034): the serverFix: Pin versions and sources; vendor anything fetched at install time.: `setup doesn't write that folder: update it the way you installed it (npx skills update), or delete it; the agents also read ${tilde(shared)}, which setup keeps current.`;Fix: Pin versions and sources; vendor anything fetched at install time.…ch, subagents, memory). ### Seeing what it did Each scenario runs `gemini -p "<prompt>" -m <model> -o stream-json --approval-mode default --allowed-mcp-server-names r…Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.