diagnosing-bugs
Diagnosis loop for a bug or slowdown whose cause is unknown. Use when asked to diagnose or debug, or something breaks, throws, fails or is slow.
- License 许可
- MIT
- 最近更新
- 2026-11-05
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- diagnosing-bugs
- description
- Diagnosis loop for a bug or slowdown whose cause is unknown. Use when asked to diagnose or debug, or something breaks, throws, fails or is slow.
02 资源文件 1
- 脚本scripts/hitl-loop.template.sh1.6 KB
03 SKILL.md 正文
<!-- Adapted from mattpocock/skills@4ae2317 skills/engineering/diagnosing-bugs, MIT (© 2026 Matt Pocock). Changed for dk-agent-kit: model-invoked with a shorter description; loops and the HITL copy live in the issue's scratch folder, and the user runs that copy in their own terminal, pasting its output back redacted as Phase 1's run; artifacts captured through the helper docs/agents/observability.md names; the regression test goes through dk:tdd and the final proof through dk:verify, whose control run carries the Phase 2 scenario, at the branch's last pre-fix commit for a bug born on the branch; the confirmed cause goes into the PR through dk:pr; a missing seam is handed to the user's /dk:improve-codebase-architecture. -->
Diagnosing Bugs
A discipline for hard bugs. Skip phases only when explicitly justified.
When exploring the codebase, read GLOSSARY.md (if it exists) to get a clear mental model of the relevant modules, and check ADRs in the area you're touching.
Loop scripts, harnesses, fixtures, captured artifacts and your copy of scripts/hitl-loop.template.sh (in this skill's base directory) live in $SCRATCH/<N>/, the issue's scratch folder from dk:issue-flow step 2.
Redact
This skill has you show commands, outputs and captured artifacts. Redact every secret first: write <REDACTED> in its place. Build loops against env vars, so the credential stays in the environment rather than in what you show. Captured artifacts carry auth headers: quote only the lines that carry the signal.
Capture production logs, error events and traces with the access command docs/agents/observability.md names, which holds the credential for you, and save them in $SCRATCH/<N>/. A repo without that doc gets the artifact from the user (Phase 1, when you cannot build a loop).
If the redacted output is not enough to diagnose the bug, say so and ask the user.
Phase 1: Build a feedback loop
This is the skill. Everything else is mechanical. If you have a tight pass/fail signal for the bug (one that goes red on this bug), you will find the cause; bisection, hypothesis-testing, and instrumentation all just consume it. If you don't have one, no amount of staring at code will save you.
Spend disproportionate effort here. Be aggressive. Be creative. Refuse to give up.
Ways to construct one, in roughly this order
- Failing test at whatever seam reaches the bug: unit, integration, e2e.
- Curl / HTTP script against a running dev server.
- CLI invocation with a fixture input, diffing stdout against a known-good snapshot.
- Headless browser script (Playwright / Puppeteer) that drives the UI and asserts on DOM/console/network.
- Replay a captured trace. Save a real network request / payload / event log to disk; replay it through the code path in isolation.
- Throwaway harness. Spin up a minimal subset of the system (one service, mocked deps) that exercises the bug code path with a single function call.
- Property / fuzz loop. If the bug is "sometimes wrong output", run 1000 random inputs and look for the failure mode.
- Bisection harness. If the bug appeared between two known states (commit, dataset, version), automate "boot at state X, check, repeat" so you can
git bisect runit. - Differential loop. Run the same input through old-version vs new-version (or two configs) and diff outputs.
- HITL bash script. Last resort. If a human must click, drive them with your copy of
scripts/hitl-loop.template.shso the loop is still structured. The user runs it in their own terminal, since your shell has none on stdin, and pastes back theKEY=VALUElines it prints.
Build the right feedback loop, and the bug is 90% fixed.
Tighten the loop
Treat the loop as a product. Once you have a loop, tighten it:
- Can I make it faster? (Cache setup, skip unrelated init, narrow the test scope.)
- Can I make the signal sharper? (Assert on the specific symptom, not "didn't crash".)
- Can I make it more deterministic? (Pin time, seed RNG, isolate filesystem, freeze network.)
A 30-second flaky loop is barely better than no loop; a 2-second deterministic one is tight, a debugging superpower.
Non-deterministic bugs
The goal is not a clean repro but a higher reproduction rate. Loop the trigger 100×, parallelise, add stress, narrow timing windows, inject sleeps. A 50%-flake bug is debuggable; 1% is not, so keep raising the rate until it's debuggable.
When you genuinely cannot build a loop
Stop and say so explicitly. List what you tried. Ask the user for: (a) access to whatever environment reproduces it, (b) a redacted captured artifact (HAR file, log dump, core dump, screen recording with timestamps), or (c) permission to add temporary production instrumentation. Do not proceed to hypothesise without a loop.
Completion criterion: a tight loop that goes red
Phase 1 is done when the loop is tight and red-capable: you can name one command (a script path, a test invocation, a curl) that you have already run at least once (show the invocation and its output, redacted); for the HITL script, the user's run counts, shown as the KEY=VALUE lines they pasted back. It is:
- [ ] Red-capable: it drives the actual bug code path and asserts the user's exact symptom, so it can go red on this bug and green once fixed. Not "runs without erroring"; it must be able to catch this specific bug.
- [ ] Deterministic: same verdict every run (flaky bugs: a pinned, high reproduction rate, per above).
- [ ] Fast: seconds, not minutes.
- [ ] Agent-runnable: you can run it unattended; a human in the loop only via
scripts/hitl-loop.template.sh, which the user runs.
If you catch yourself reading code to build a theory before this command exists, stop: jumping straight to a hypothesis is the exact failure this skill prevents. No red-capable command, no Phase 2.
Phase 2: Reproduce + minimise
Run the loop. Watch it go red as the bug appears.
Confirm:
- [ ] The loop produces the failure mode the user described, not a different failure that happens to be nearby. Wrong bug = wrong fix.
- [ ] The failure is reproducible across multiple runs (or, for non-deterministic bugs, reproducible at a high enough rate to debug against).
- [ ] You have captured the exact symptom (error message, wrong output, slow timing) so later phases can verify the fix actually addresses it.
Minimise
Once it's red, shrink the repro to the smallest scenario that still goes red. Cut inputs, callers, config, data, and steps one at a time, re-running the loop after each cut, and keep only what's load-bearing for the failure.
Why bother: a minimal repro shrinks the hypothesis space in Phase 3 (fewer moving parts left to suspect) and becomes the clean regression test in Phase 5.
Done when every remaining element is load-bearing: removing any one of them makes the loop go green.
Do not proceed until you have reproduced and minimised.
Keep the un-minimised loop and its red output: they are the scenario and the expected control result for dk:verify in Phase 6.
Phase 3: Hypothesise
Generate 3–5 ranked hypotheses before testing any of them. Single-hypothesis generation anchors on the first plausible idea.
Each hypothesis must be falsifiable: state the prediction it makes.
Format: "If <X> is the cause, then <changing Y> will make the bug disappear / <changing Z> will make it worse."
If you cannot state the prediction, the hypothesis is a vibe: discard or sharpen it.
Show the ranked list to the user before testing. They often have domain knowledge that re-ranks instantly ("we just deployed a change to #3"), or know hypotheses they've already ruled out. Cheap checkpoint, big time saver. Don't block on it; proceed with your ranking if the user is AFK.
Phase 4: Instrument
Each probe must map to a specific prediction from Phase 3. Change one variable at a time.
Tool preference:
- Debugger / REPL inspection if the env supports it. One breakpoint beats ten logs.
- Targeted logs at the boundaries that distinguish hypotheses.
- Never "log everything and grep".
Tag every debug log with a unique prefix, e.g. [DEBUG-a4f2]. Cleanup at the end becomes a single grep. Untagged logs survive; tagged logs die.
Perf branch. For performance regressions, logs are usually wrong. Instead: establish a baseline measurement (timing harness, performance.now(), profiler, query plan), then bisect. Measure first, fix second.
Phase 5: Fix + regression test
Write the regression test before the fix, but only if there is a correct seam for it.
A correct seam is one where the test exercises the real bug pattern as it occurs at the call site. If the only available seam is too shallow (single-caller test when the bug needs multiple callers, unit test that can't replicate the chain that triggered the bug), a regression test there gives false confidence.
If no correct seam exists, that itself is the finding. The codebase architecture is preventing the bug from being locked down. State it in your report and in the PR, and recommend the user run /dk:improve-codebase-architecture to find the seam; it is user-invoked, so starting it is theirs.
If a correct seam exists, call the Skill tool with dk:tdd, propose that seam under its seam rule, and run its loop through these steps:
- Turn the minimised repro into a failing test at that seam.
- Watch it fail. If you forced the red by mutating code or a fixture,
diffagainst a pristine copy to prove the mutation landed before you trust it. - Apply the fix.
- Watch it pass.
- Re-run the Phase 1 feedback loop against the original (un-minimised) scenario.
Phase 6: Cleanup
Required before declaring done:
- [ ] Original repro no longer reproduces (re-run the Phase 1 loop)
- [ ] Regression test passes (or absence of seam is documented)
- [ ] All
[DEBUG-...]instrumentation removed (grepthe prefix) - [ ] Throwaway harnesses and scripts sit in
$SCRATCH/<N>/, andgit statusshows none in the tree - [ ] The fix is proven: call the Skill tool with
dk:verify, with the loop kept in Phase 2 as its control run's scenario, and the branch's last pre-fix commit as the control when the bug came in on this branch (dk:verifystep 4) - [ ] The hypothesis that turned out correct is stated in the commit message, and in the PR body you write after calling the Skill tool with
dk:pr, so the next debugger learns
04 同仓库更多 Skills
writing-for-agents
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
verify
Prove a change works by driving it in a stack you started, against a control run of the pre-fix state. Use before calling any change verified, or when asked to verify or prove a fix.
verify-doc-check
Monthly check that this repo's docs/agents/verify.md still starts the stack: read every claim against the source, run the recipe in a fresh worktree, fix or file each drifted step.
triage
Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.
to-tickets
Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker (edges as text in one file per ticket locally, or native blocking links on a real tracker).
to-spec
Turn the current conversation into a spec and publish it to the project issue tracker: no interview, just synthesis of what you've already discussed.