hydra-nano@0.3.3
CAUTION· install onlyPublisher claims: "Small local models that take the noisy work off your coding agent: git history, test and build digests and OS inspection run on your machine, and only the lines that matter reach Claude Code, Codex, Cursor, OpenCode, Gemini CLI or Antigravity. Skill, MCP " Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Bundled scripts ran with no arguments; no agent task scenario for this bundle.- Tested
- 8 of 21 bundled scripts run with no arguments (the rest were not run: per-bundle cap), with decoy credentials in the environment · every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
What changed since 0.3.1
vet-skills diff hydra-nano 0.3.1 0.3.3This version reaches for something 0.3.1 did not: 2 new hosts (ollama.com, refactoring.guru). Compared from the two recorded runs, not from source.
- New hosts
- ollama.com, refactoring.guru
The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-20 and scored it 78/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Access to sensitive file paths. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 1 s · 2 scenariosStatic findings
…ND=ollama`; its `setup` downloads models, so tell the user before starting it. `npx skills add` installs skill/CLI files only; `npx github:rahul05ranjan/hydra` performs t…Fix: Pin versions and sources; vendor anything fetched at install time.// A pinned skills CLI: `npx --yes` would otherwise run whatever version is newest on the registry today.Fix: Pin versions and sources; vendor anything fetched at install time./** Copy the skill's scripts to sharedScripts(), so registrations never point into an npx cache.Fix: Pin versions and sources; vendor anything fetched at install time.const npx = findNpx();Fix: Pin versions and sources; vendor anything fetched at install time.if (!existsSync(source)) throw new Manual(dir, 'npx github:rahul05ranjan/hydra', `needs the hydra skill, but ${source} is missing; run the installer again`);Fix: Pin versions and sources; vendor anything fetched at install time.writeFileSync(temp, JSON.stringify(config, null, 2) + '\n');Fix: Declare and request only the permissions the task needs.{ cwd, env: { ...process.env, ...env }, maxBuffer: 32 * 1024 * 1024, windowsHide: true },Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.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
No advisories in OSV for hydra-nano@0.3.3 as of 1 min ago. This is the package itself, not its dependencies.
What would make this Verified
2 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim rahul05ranjan: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Touches files outside the project≈ +7 on the totalKeep reads and writes inside the working directory, or declare the exact paths in `filesystem:` and explain why in the README.
- Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.
- Says it only reads; the code writes or deletes files≈ +2 on the totalDeclare and request only the permissions the task needs.