apply-findings

Name: apply-findings
Author: tobihagemann

3installs29forks

How to install apply-findings?

tobihagemann/turbo/apply-findings

$npx -y skills add tobihagemann/turbo --skill apply-findings

Installs into the current project.

Summary

Apply findings by making the suggested code changes. Applies accepted verdicts, escalates ambiguous findings to the user, and offers to note genuine improvements for later. Use when the user asks to \"apply findings\", \"apply fixes\", \"apply suggestions\", \"apply accepted find

Files · 1

View on GitHub

SKILL.md

1	# Apply Findings
2
3	Apply evaluated findings from the conversation context. Findings must have been through `/evaluate-findings` first.
4
5	## Step 1: Identify Findings
6
7	Collect all findings from the conversation context. Findings should have Verdict columns (Apply, Skip, Escalate) from `/evaluate-findings`.
8
9	If findings are unevaluated (raw output without verdicts), stop and say to run `/evaluate-findings` first.
10
11	## Step 2: Apply in File Order
12
13	Group Apply findings by file path and apply in file order to minimize context switching. For each finding:
14
15	1. Read the full function or logical block at the referenced location
16	2. Verify the finding still applies to the current code
17	3. When the finding carries a suggested fix, treat the fix as a separate claim from the finding and verify it independently before applying it — trace it against the failure modes the finding names. If the fix does not hold up, treat the finding as Escalate (surface it in Step 3) and record why the remedy fails, rather than applying an unsound fix on the finding's authority.
18	4. Make the fix
19	5. If the finding renames an identifier, search the file for all occurrences of the old name before marking the fix complete. The cited location is often only one of several references.
20
21	If a finding references code that has changed since it was generated (e.g., by a prior fix in this same run), re-assess whether it still applies. Skip if the code has diverged.
22
23	## Step 3: Handle Escalated Findings
24
25	For findings with Escalate verdict, use `AskUserQuestion` to let the user decide. Recommend the genuinely best option: place it first and append `(Recommended)` to its label, judging "best" on technical merit alone (the soundest engineering outcome), independent of how closely the option conforms to the task's original scope. When the choice hinges on product intent or domain knowledge you lack and merit cannot settle it, say so instead of forcing a pick. Give each option a plain-language description that carries the trade-off: its concrete effect and what it costs. When the recommended option also widens the changeset's scope, name both its merit and that scope cost so the user can weigh them.
26
27	- Apply — make the change
28	- Skip — leave as-is
29	- Note for later — run the `/note-improvement` skill to capture it
30
31	## Step 4: Report Results
32
33	Summarize what was applied, what was escalated, and what was skipped.
34
35	Then use the TaskList tool and proceed to any remaining task.
36
37	## Rules
38
39	- Only edit files. Do not stage, build, or test.
40	- If two Apply findings conflict (suggest opposite changes to the same code), surface the conflict with `AskUserQuestion`, recommend the genuinely best option on technical merit, and let the user choose, rather than applying either.

People who liked this also viewed

6 picks

Type

apply-findings

3installs29forks

View on GitHub

How to install apply-findings?

tobihagemann/turbo/apply-findings

$npx -y skills add tobihagemann/turbo --skill apply-findings

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Apply Findings
2
3	Apply evaluated findings from the conversation context. Findings must have been through `/evaluate-findings` first.
4
5	## Step 1: Identify Findings
6
7	Collect all findings from the conversation context. Findings should have Verdict columns (Apply, Skip, Escalate) from `/evaluate-findings`.
8
9	If findings are unevaluated (raw output without verdicts), stop and say to run `/evaluate-findings` first.
10
11	## Step 2: Apply in File Order
12
13	Group Apply findings by file path and apply in file order to minimize context switching. For each finding:
14
15	1. Read the full function or logical block at the referenced location
16	2. Verify the finding still applies to the current code
17	3. When the finding carries a suggested fix, treat the fix as a separate claim from the finding and verify it independently before applying it — trace it against the failure modes the finding names. If the fix does not hold up, treat the finding as Escalate (surface it in Step 3) and record why the remedy fails, rather than applying an unsound fix on the finding's authority.
18	4. Make the fix
19	5. If the finding renames an identifier, search the file for all occurrences of the old name before marking the fix complete. The cited location is often only one of several references.
20
21	If a finding references code that has changed since it was generated (e.g., by a prior fix in this same run), re-assess whether it still applies. Skip if the code has diverged.
22
23	## Step 3: Handle Escalated Findings
24
25	For findings with Escalate verdict, use `AskUserQuestion` to let the user decide. Recommend the genuinely best option: place it first and append `(Recommended)` to its label, judging "best" on technical merit alone (the soundest engineering outcome), independent of how closely the option conforms to the task's original scope. When the choice hinges on product intent or domain knowledge you lack and merit cannot settle it, say so instead of forcing a pick. Give each option a plain-language description that carries the trade-off: its concrete effect and what it costs. When the recommended option also widens the changeset's scope, name both its merit and that scope cost so the user can weigh them.
26
27	- Apply — make the change
28	- Skip — leave as-is
29	- Note for later — run the `/note-improvement` skill to capture it
30
31	## Step 4: Report Results
32
33	Summarize what was applied, what was escalated, and what was skipped.
34
35	Then use the TaskList tool and proceed to any remaining task.
36
37	## Rules
38
39	- Only edit files. Do not stage, build, or test.
40	- If two Apply findings conflict (suggest opposite changes to the same code), surface the conflict with `AskUserQuestion`, recommend the genuinely best option on technical merit, and let the user choose, rather than applying either.

People who liked this also viewed

6 picks

Type