How do I install the Pflow Golang Troubleshoot skill?

Run npx skillmds@latest add phpinfo/pflow-golang-troubleshoot in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Pflow Golang Troubleshoot safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Pflow Golang Troubleshoot?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Pflow Golang Troubleshoot free to use?

Yes. Installing skills from SkillMD is free. This skill is licensed under MIT.

Who published Pflow Golang Troubleshoot?

phpinfo (@phpinfo) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Pflow Golang Troubleshoot by phpinfo

Name: Pflow Golang Troubleshoot
Author: phpinfo

Pflow Golang Troubleshoot

Systematic root-cause debugging for Go programs — panics, wrong results, failing or flaky tests, deadlocks, data races, goroutine leaks, memory or CPU blow-ups, build and module errors. Detects the project's toolchain, debug tools and test entry points, then guides the investigation with pprof, the race detector, delve and targeted tests. Use when Go code misbehaves and the cause is unknown. Not for writing new code (pflow-golang).

phpinfo Updated Sep 17, 2026

File contents

If troubleshoot-context.sh fails (non-zero exit or "status":"error" in its JSON) print ⚠️ <error.message> and stop. Other failing commands are evidence — record and continue.

Steps

Run .agents/skills/pflow-golang-troubleshoot/scripts/troubleshoot-context.sh → {toolchain, goos, goarch, cgo_enabled, debug_tools[], task_runner, test_targets[], test_files, test_dirs[], test_build_tags, hint}.

Read references/methodology.md, then the file for the symptom:

Symptom	Read
panic, wrong output, nil pointer, index out of range	`common-bugs.md`
hangs, deadlock, race report, leak, "all goroutines are asleep"	`concurrency-debug.md`
slow, high memory, high CPU, GC pressure	`pprof-delve.md`
test fails, flaky, passes alone but fails together, coverage gaps	`testing-debug.md`
does not compile, `go mod` / vet / lint errors	`build-errors.md`

Reproduce first, with the narrowest command available: a test_targets entry from the project runner, else go test -run 'TestX' -race -count=1 ./pkg/.... Respect test_build_tags (e.g. -tags e2e needs its backend — ask before running).
Form one hypothesis at a time, prove it with evidence (log, test, profile, dlv), fix the root cause, add a regression test, rerun the reproduction and the project gate.
Report: symptom → root cause → fix → proof (test name / command output). Never report "fixed" without the reproduction passing.

Paths are relative to .agents/skills/pflow-golang-troubleshoot/.

Gotchas

hint non-null means CGO_ENABLED=0; -race may need CGO_ENABLED=1 go test -race.
Missing dlv/benchstat/govulncheck in debug_tools: propose go install …@latest or go run …@latest; don't assume they exist.
Don't change behavior to make a symptom disappear; if the fix widens beyond the cause, say so and stop for confirmation.

---name: pflow-golang-troubleshootdescription: Systematic root-cause debugging for Go programs — panics, wrong results, failing or flaky tests, deadlocks, data races, goroutine leaks, memory or CPU blow-ups, build and module errors. Detects the project's toolchain, debug tools and test entry points, then guides the investigation with pprof, the race detector, delve and targeted tests. Use when Go code misbehaves and the cause is unknown. Not for writing new code (pflow-golang).license: MIT---If `troubleshoot-context.sh` fails (non-zero exit or `"status":"error"` in its JSON) print `⚠️ <error.message>` and stop. Other failing commands are evidence — record and continue.## Steps1. Run `.agents/skills/pflow-golang-troubleshoot/scripts/troubleshoot-context.sh` → `{toolchain, goos, goarch, cgo_enabled, debug_tools[], task_runner, test_targets[], test_files, test_dirs[], test_build_tags, hint}`.2. Read `references/methodology.md`, then the file for the symptom: | Symptom | Read | | --- | --- | | panic, wrong output, nil pointer, index out of range | `common-bugs.md` | | hangs, deadlock, race report, leak, "all goroutines are asleep" | `concurrency-debug.md` | | slow, high memory, high CPU, GC pressure | `pprof-delve.md` | | test fails, flaky, passes alone but fails together, coverage gaps | `testing-debug.md` | | does not compile, `go mod` / vet / lint errors | `build-errors.md` |3. Reproduce first, with the narrowest command available: a `test_targets` entry from the project runner, else `go test -run 'TestX' -race -count=1 ./pkg/...`. Respect `test_build_tags` (e.g. `-tags e2e` needs its backend — ask before running).4. Form one hypothesis at a time, prove it with evidence (log, test, profile, `dlv`), fix the root cause, add a regression test, rerun the reproduction and the project gate.5. Report: symptom → root cause → fix → proof (test name / command output). Never report "fixed" without the reproduction passing.Paths are relative to `.agents/skills/pflow-golang-troubleshoot/`.## Gotchas- `hint` non-null means `CGO_ENABLED=0`; `-race` may need `CGO_ENABLED=1 go test -race`.- Missing `dlv`/`benchstat`/`govulncheck` in `debug_tools`: propose `go install …@latest` or `go run …@latest`; don't assume they exist.- Don't change behavior to make a symptom disappear; if the fix widens beyond the cause, say so and stop for confirmation.

Frequently asked questions