diagnose

Name: diagnose
Author: avibebuilder

109stars17forks

How to install diagnose?

avibebuilder/claude-prime/diagnose

$npx -y skills add avibebuilder/claude-prime --skill diagnose

Installs into the current project.

Summary

Investigate unexpected behavior and mysterious bugs. Use when the cause of a problem is unknown and the user needs to understand WHY something is happening — symptoms like: sudden unexplained changes in metrics or behavior, works locally but not in staging/production, inconsisten

Files · 1

View on GitHub

SKILL.md

1	Think harder.
2
3	## Process
4
5	Check conversation context and skip completed steps.
6
7	### 1. Understand the symptom
8	- Read the bug report, errors, logs, and surrounding code carefully
9	- Clarify reproduction steps, expected behavior, and environment when they are unclear
10	- Separate confirmed facts from working assumptions. List them explicitly:
11	- `Fact (confirmed):` the server returns 200
12	- `Assumption (unconfirmed):` the client receives the full HTML body
13	Misidentifying an assumption as a fact is the most common source of wasted investigation.
14
15	### 2. Build hypotheses
16	- Form 2-4 plausible root-cause hypotheses that are mechanistically distinct — different failure layers (e.g., server render vs. client hydration vs. network layer), not variations of the same idea
17	- Rank them by likelihood
18	- For each hypothesis, state both sides:
19	- `Confirm if:` [what observation would prove this is the cause]
20	- `Eliminate if:` [what observation would rule this out]
21
22	A hypothesis you can't falsify in both directions is too vague to test.
23
24	### 3. Choose the lightest evidence method
25	Start with the cheapest source of truth that can kill hypotheses:
26	- existing logs, traces, stack traces, metrics, and error output
27	- static code inspection around the suspected path
28	- config, environment, deploy, cache, queue, and permissions state that could explain the symptom
29	- targeted reproduction in the relevant environment
30
31	Only add new instrumentation when existing evidence is insufficient.
32	- If you need runtime probes, read `diagnose/references/runtime-debugging.md`
33	- Use `#region agent log` / `#endregion` markers for any instrumentation you add
34	- Tag each log point with the relevant `hypothesisId`
35	- Log only the minimum fields needed to discriminate between hypotheses; never log secrets, tokens, passwords, cookies, or full sensitive payloads
36	- If runtime probes require starting the local debug server, ask the user before launching it
37	- For browser/UI bugs, combine with the `agent-browser` skill when reproduction or inspection needs it
38
39	### 4. Gather evidence and iterate
40	- Use existing logs, traces, failing tests, or artifacts before asking for a fresh reproduction
41	- When reproduction is needed, ask the user to trigger the bug — tie each request to the hypothesis it tests
42	- Correlate each finding with the hypothesis it supports or eliminates; narrow based on evidence, not confidence
43	- If ambiguity remains, refine hypotheses and add narrower probes — but stop and report when another round is unlikely to produce new discriminating evidence
44
45	### 5. Report the diagnosis
46
47	Output structured diagnosis:
48
49	```
50	## Diagnosis: [Issue Title]
51
52	### Symptoms
53	- [What was observed]
54
55	### Evidence
56	- [Finding] — `file:line` or runtime source — hypothesis X
57	- ...
58
59	### Root Cause
60	[Confirmed or most likely cause, with evidence]
61
62	### Hypotheses Tested
63	\| # \| Hypothesis \| Confirm if \| Eliminate if \| Result \|
64	\|---\|-----------\|-----------\|-------------\|--------\|
65	\| A \| ... \| [what observation would prove this] \| [what observation would rule this out] \| Confirmed/Eliminated/Inconclusive \|
66
67	### Recommended Next Steps
68	- [What to do next — usually hand off to `/fix` with this diagnosis]
69
70	### Active Instrumentation
71	- [List files with `#region agent log` blocks still in place, or `None`]
72	```
73
74	## Constraints
75
76	- NO fixing — investigation and diagnosis only. "Recommended Next Steps" hands off to `/fix` with the diagnosis; it does not prescribe specific parameter values, code snippets, or step-by-step implementation instructions.
77	- Evidence over assumptions — if the code looks wrong but runtime evidence says otherwise, trust the runtime
78	- If you add `#region agent log` blocks, leave them in place for `/fix` to verify the repair and call them out in the final report
79
80	## Bug
81
82	<bug>$ARGUMENTS</bug>

People who liked this also viewed

6 picks

Type

diagnose

109stars17forks

View on GitHub

How to install diagnose?

avibebuilder/claude-prime/diagnose

$npx -y skills add avibebuilder/claude-prime --skill diagnose

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	Think harder.
2
3	## Process
4
5	Check conversation context and skip completed steps.
6
7	### 1. Understand the symptom
8	- Read the bug report, errors, logs, and surrounding code carefully
9	- Clarify reproduction steps, expected behavior, and environment when they are unclear
10	- Separate confirmed facts from working assumptions. List them explicitly:
11	- `Fact (confirmed):` the server returns 200
12	- `Assumption (unconfirmed):` the client receives the full HTML body
13	Misidentifying an assumption as a fact is the most common source of wasted investigation.
14
15	### 2. Build hypotheses
16	- Form 2-4 plausible root-cause hypotheses that are mechanistically distinct — different failure layers (e.g., server render vs. client hydration vs. network layer), not variations of the same idea
17	- Rank them by likelihood
18	- For each hypothesis, state both sides:
19	- `Confirm if:` [what observation would prove this is the cause]
20	- `Eliminate if:` [what observation would rule this out]
21
22	A hypothesis you can't falsify in both directions is too vague to test.
23
24	### 3. Choose the lightest evidence method
25	Start with the cheapest source of truth that can kill hypotheses:
26	- existing logs, traces, stack traces, metrics, and error output
27	- static code inspection around the suspected path
28	- config, environment, deploy, cache, queue, and permissions state that could explain the symptom
29	- targeted reproduction in the relevant environment
30
31	Only add new instrumentation when existing evidence is insufficient.
32	- If you need runtime probes, read `diagnose/references/runtime-debugging.md`
33	- Use `#region agent log` / `#endregion` markers for any instrumentation you add
34	- Tag each log point with the relevant `hypothesisId`
35	- Log only the minimum fields needed to discriminate between hypotheses; never log secrets, tokens, passwords, cookies, or full sensitive payloads
36	- If runtime probes require starting the local debug server, ask the user before launching it
37	- For browser/UI bugs, combine with the `agent-browser` skill when reproduction or inspection needs it
38
39	### 4. Gather evidence and iterate
40	- Use existing logs, traces, failing tests, or artifacts before asking for a fresh reproduction
41	- When reproduction is needed, ask the user to trigger the bug — tie each request to the hypothesis it tests
42	- Correlate each finding with the hypothesis it supports or eliminates; narrow based on evidence, not confidence
43	- If ambiguity remains, refine hypotheses and add narrower probes — but stop and report when another round is unlikely to produce new discriminating evidence
44
45	### 5. Report the diagnosis
46
47	Output structured diagnosis:
48
49	```
50	## Diagnosis: [Issue Title]
51
52	### Symptoms
53	- [What was observed]
54
55	### Evidence
56	- [Finding] — `file:line` or runtime source — hypothesis X
57	- ...
58
59	### Root Cause
60	[Confirmed or most likely cause, with evidence]
61
62	### Hypotheses Tested
63	\| # \| Hypothesis \| Confirm if \| Eliminate if \| Result \|
64	\|---\|-----------\|-----------\|-------------\|--------\|
65	\| A \| ... \| [what observation would prove this] \| [what observation would rule this out] \| Confirmed/Eliminated/Inconclusive \|
66
67	### Recommended Next Steps
68	- [What to do next — usually hand off to `/fix` with this diagnosis]
69
70	### Active Instrumentation
71	- [List files with `#region agent log` blocks still in place, or `None`]
72	```
73
74	## Constraints
75
76	- NO fixing — investigation and diagnosis only. "Recommended Next Steps" hands off to `/fix` with the diagnosis; it does not prescribe specific parameter values, code snippets, or step-by-step implementation instructions.
77	- Evidence over assumptions — if the code looks wrong but runtime evidence says otherwise, trust the runtime
78	- If you add `#region agent log` blocks, leave them in place for `/fix` to verify the repair and call them out in the final report
79
80	## Bug
81
82	<bug>$ARGUMENTS</bug>

People who liked this also viewed

6 picks

Type