More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    gopolice

    1

    One-command Go code audit (lint, security, tests, dead code, complexity, benchmarks) with a prioritised fix loop.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    gopolice

    gopolice

    Example session with this skill installed

    Use my Agensi skill "gopolice" to audit this project. Give me the grade and the top security issues.

    • Read your context and instructions
    • Compiled the gopolice

    I ran gopolice on docker/cli (728 Go files, 110k lines). Here is the audit.

    Grade: F (794 points). D starts at 80, so this is a long way off, but the score is dominated by two things: 13 stdlib vulnerabilities and 181 gosec findings, most of them one rule.

    Issues: 295 total (29 errors, 119 warnings, 147 info)
    govulncheck 13 · gosec 181 · golangci-lint 11 · dead code 2 · complexity 88

    Tests: 2672 passed, 0 failed, 5 skipped. Coverage averages 74.7% over 64 tested packages. Lowest: cli-plugins/plugin 2.7%, internal/jsonstream 10%, cli/connhelper 17.6%.

    Top security issues

    1. 13 known vulnerabilities in the Go standard library (net/http, crypto/tls, crypto/x509, html/template, mime/multipart). All are fixed in Go 1.27.2, one in 1.27.3. Updating the toolchain in go.mod clears every one of them. This is the single highest-value fix.
    2. Command injection via taint analysis (gosec G702) in cli-plugins/manager/manager.go:213. An exec call built from values that reach it from outside. Needs a manual review of where the arguments come from.
    3. Potential hardcoded credentials (G101) in cli/command/service/opts.go:985 and cli/config/credentials/native_store.go:10. Likely constant names that look like secrets, but confirm before dismissing.
    4. Race-prone filesystem operation inside filepath.Walk (G122) in cli/command/image/build/context.go:79. Symlink TOCTOU; os.Root or a re-check after Lstat closes it.
    5. Integer overflow conversions (G115), about a dozen as errors across cli/command/service/progress, container/opts.go and stats_helpers.go. Each is a uint64 to int cast on sizes from the API; bounds-check or widen the target type.

    What to fix first

    1. Bump the Go toolchain to 1.27.3. Removes 13 errors in one change.
    2. Review the G702 exec call and the two G101 hits.
    3. Delete the two unused functions staticcheck found, (*Writer).dump in tabwriter.go and (*dummy).func2 in reflect_test.go, unless the test fixture is deliberate.
    4. Work through the G115 casts.
    5. The 88 complexity warnings and 11 lint infos can wait; none block the grade until the errors are gone.

    The full report is in gopolice.json. Say the word and I will apply fixes 1 to 3 and rescan so you can see the before and after grade, or run gopolice serve to open the dashboard.

    The full report is in gopolice.json

    Connects securely to your tools. The creator never sees your data.

    About this skill

    gopolice turns any Go repository into a graded, prioritised fix list your agent can work through.

    One command runs go vet, golangci-lint, gosec, govulncheck, staticcheck, go test with coverage, benchmarks and CPU/memory profiling, then merges everything into a single JSON report with a quality grade from A to F. The bundled triage script reads that report and prints what to fix first: tool gaps, failing tests, known vulnerabilities, security findings, vet and lint errors, dead code, then complexity. Grade, coverage by package and CPU hot spots fit on one screen.

    What the skill teaches your agent

    • Install gopolice with go install if it is missing, and report which optional scanners are absent instead of guessing.
    • Run a headless scan that writes one report and never opens a browser or leaves a server running.
    • Triage the report in priority order and explain each finding with its rule, location and suggested fix.
    • Make small, behaviour-preserving fixes, rescan, and diff the two reports so you see exactly what was fixed, what is new, and the grade before and after.
    • Export SARIF for GitHub code scanning or CI.
    • Handle go.work workspaces, with issues grouped per module.
    • Hand the result to a human: gopolice serve opens the cached report in a local dashboard with issues, security, tests, performance, dead code, dependencies and git views.

    Good for

    • Auditing a Go repository before a release and fixing what matters first.
    • Finding failing tests, low coverage and dead code across a monorepo.
    • Explaining why a project's grade dropped and bringing it back up.
    • Producing a SARIF report from a Go project.

    Requirements and limits

    • Go 1.26 or newer on PATH. The agent installs gopolice itself; golangci-lint, gosec, govulncheck and staticcheck are optional and add scanners when present.
    • Scans execute the project's tests and benchmarks, so expect a few minutes on large repos.
    • govulncheck needs network access to fetch the vulnerability database.
    • gopolice reports; it does not auto-fix. Your agent makes the edits and verifies them with a rescan.
    • Go projects only.

    gopolice is open source (MIT) at github.com/rapando/gopolice.

    Requires Go 1.26+. Scans execute code via tests/benchmarks. govulncheck needs network access. Does not auto-fix code.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 12 days ago

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Listed12 days ago

    What's inside

    Frequently Asked Questions

    gopolice automates the orchestration of multiple Go tools like gosec and staticcheck, replacing a manual, fragmented audit with a single prioritized report and a grade-based quality score.

    This skill is compatible with any AI agent capable of executing terminal commands and requires Go 1.26 or newer installed on the host environment.

    No, the skill performs a 'headless' scan by default, but it will automatically utilize tools like golangci-lint and gosec if they are present in your environment for a deeper analysis.

    Yes, gopolice includes a regression tracking feature that compares your current scan against a baseline to verify that fixes were effective and that no new issues were introduced.

    Because gopolice executes benchmarks and 'go test' to verify code health, it should only be used on trusted source code within a secure environment.

    New in Code Quality & Review