change-report-delivery
Skill0 starsChanged 3 days agoMITView on GitHub
Development change report and delivery standard. Requires truthfully describing added, modified, and deleted files, functions, logic, interfaces, and verification results.
In this repository17 other files it gives its agents.
- .cursor/rules/engineering-context.mdcCursor rule
- code-asset-inventorySkill
- data-audit-reproducibilitySkill
- data-first-executionSkill
- documentation-structureSkill
- engineering-context-completenessSkill
- engineering-transparencySkill
- explicit-instruction-fidelitySkill
- interface-contractSkill
- language-expressionSkill
- production-loggingSkill
- repository-publication-hygieneSkill
- task-execution-governanceSkill
- technical-documentationSkill
- testing-verificationSkill
- unfinished-work-trackingSkill
- user-intent-prioritySkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
After every development, modification, refactoring, fix, or configuration adjustment is complete, a delivery description consistent with …
- Distinctive: line 33
Only report changes that actually occurred.
- Distinctive: line 106
When data processing is involved, state:
- Distinctive: line 123
When module collaboration is involved, state:
- Distinctive: line 131
Use a flow diagram where necessary.
- Distinctive: line 187
Prefer Git Diff or other version diff tools.
- Distinctive: line 193
Test results must not be fabricated.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Development Change Report and Delivery Standard
I. Core Principle
After every development, modification, refactoring, fix, or configuration adjustment is complete, a delivery description consistent with the actual changes must be provided.
The user should not have to re-read all the source code in order to know what was done this time.
The following statements must not be used on their own as a delivery report:
- Development completed.
- Logic optimized.
- Issue fixed.
- Files updated.
- Testing completed.
II. Overview of This Work
State:
- The original objective.
- The problem solved.
- The work actually completed.
- The final effect.
- The content that is not yet complete.
Only report changes that actually occurred.
III. File Change List
Divided into three categories.
1. Added Files
State:
- Complete path.
- File type.
- File purpose.
- Main responsibilities.
- Inputs and outputs.
- Related modules.
2. Modified Files
State:
- Complete path.
- Original responsibilities.
- Reason for the modification.
- Specific modification content.
- Responsibilities after the modification.
- Impact on other modules.
3. Deleted Files
State:
- Complete path.
- Original responsibilities.
- Reason for deletion.
- Whether functionality was migrated.
- Whether callers exist.
- Whether other modules are affected.
IV. Function-Level Changes
Added Functions
State:
- Function name.
- Containing file.
- Functionality.
- Input parameters.
- Return value.
- Core logic.
- Call relationships.
Modified Functions
State:
- Function name.
- Behavior before the modification.
- Reason for the modification.
- Specific logic changes.
- Behavior after the modification.
- Whether inputs and outputs changed.
- Impact on callers.
Deleted Functions
State:
- Function name.
- Original functionality.
- Reason for deletion.
- Whether an alternative implementation exists.
- Scope of impact.
It is not required to relist all functions that did not change.
V. Data and Processing Changes
When data processing is involved, state:
- Which data was used.
- Data source.
- Data location.
- Data structure.
- Input count.
- Processing rules.
- Filtering conditions.
- Output count.
- Final result.
- Output location.
Raw data and processing records must be independently inspectable.
VI. Call Relationship Changes
When module collaboration is involved, state:
- The original call relationships.
- The location of this change.
- The new call relationships.
- How data is passed.
- Whether upstream and downstream have changed.
Use a flow diagram where necessary.
VII. Interfaces and Compatibility
Must make clear:
- Whether interfaces were added.
- Whether interfaces were modified.
- Whether interfaces were deleted.
- Whether the input structure changed.
- Whether the output structure changed.
- Whether fields changed.
- Whether field types changed.
- Whether the invocation method changed.
- Which modules are affected.
- Whether data migration is needed.
Even when there is no interface change, this must be stated explicitly.
VIII. Testing and Verification
Must distinguish:
- Passed.
- Failed.
- Not executed.
- Unable to execute.
- Not applicable.
State the actual test commands, test scope, results, and failure reasons.
Tests that were not executed must not be written as passed.
IX. Outstanding Items
List:
- Unimplemented functionality.
- Unfixed problems.
- Unfinished documentation.
- Unverified parts.
- Matters requiring further investigation.
Record them in the project's outstanding items list as well.
X. Authenticity Requirements
The delivery report must be based on:
- Actual code differences.
- Actual file changes.
- Actual function definitions.
- Actual runtime results.
- Actual test results.
- Actual logs.
Prefer Git Diff or other version diff tools.
Existing functionality must not be described as newly added this time.
Planned modifications must not be described as already modified.
Test results must not be fabricated.
XI. Adjusting Length According to Scale
Simple modifications may be described concisely.
Module-level development requires file and function descriptions.
Cross-module refactoring requires complete descriptions of data flow, interfaces, and compatibility.
The same information must not be repeated in order to satisfy a template.
XII. Final Principle
The user must clearly know:
What was changed, why it was changed, how exactly it was changed, what impact it produced, and what remains unfinished.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

