decision-memory
Skill0 starsChanged 3 months agoApache-2.0View on GitHub
Preserve important project decisions so future AI sessions and developers do not silently contradict previous architectural choices. Inspects existing documentation and code evidence to identify decisions, then creates or updates PROJECT_DECISIONS.md with structured decision records. Distinguishes DECISIONS from PREFERENCES from TEMPORARY DETAILS. Never invents historical decisions. Use when the user wants to document architectural choices, preserve context across sessions, or prevent future agents from undoing intentional decisions.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- anuragsarkar97/skillstheir own version, first seen Nov 8, 2026
- gootf/software-agent-soptheir own version, first seen Aug 30, 2026
- weilinweng/decision-memorytheir own version, first seen Aug 8, 2026
- ltorresu82/skillstheir own version, first seen Jun 1, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- gootf/software-agent-sop★ 2same name
- anuragsarkar97/skills★ 0same name
- ltorresu82/skills★ 0same name
- weilinweng/decision-memory★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository49 other files it gives its agents.
- accessibility-auditorSkill
- api-doctorSkill
- browser-compatibilitySkill
- build-doctorSkill
- bundle-doctorSkill
- cache-doctorSkill
- change-cartographerSkill
- ci-cd-doctorSkill
- code-auditorSkill
- code-migratorSkill
- code-reviewerSkill
- config-doctorSkill
- container-doctorSkill
- context-keeperSkill
- contract-testerSkill
- cron-doctorSkill
- database-doctorSkill
- dependency-doctorSkill
- documentation-engineerSkill
- env-guardSkill
- error-boundary-doctorSkill
- feature-integritySkill
- git-hygieneSkill
- handoff-engineerSkill
- i18n-auditorSkill
- incident-debuggerSkill
- intent-architectSkill
- log-doctorSkill
- motion-designerSkill
- observability-engineerSkill
- performance-optimizerSkill
- product-architectSkill
- project-driftSkill
- qa-engineerSkill
- refactor-engineerSkill
- release-notesSkill
- runtime-doctorSkill
- safe-scopeSkill
- security-auditorSkill
- seo-doctorSkill
- ship-itSkill
- skill-managerSkill
- spec-doctorSkill
- state-managerSkill
- terminal-visualizerSkill
- test-engineerSkill
- type-safety-auditorSkill
- ui-redesignSkill
- ux-friction-hunterSkill
The file
We split this skill into the rules it gives an agent
11 rules. 0 are common, 9 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
9
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 34
Before creating anything, find what already exists:
- Very specific: line 36
Check for PROJECT_DECISIONS.md, docs/decisions/, ADR/, architecture/ directories.
- Very specific: line 37
Read README.md, CLAUDE.md, PROJECT_CONTEXT.md for embedded decisions.
- Distinctive: line 38
Check git history for commit messages that describe decisions.
- Distinctive: line 39
Look for inline code comments that explain "why" a choice was made.
- Distinctive: line 67
Only record DECISIONS and significant PREFERENCES in the decision file.
- Distinctive: line 71
If the file does not exist, create it with this structure:
- Distinctive: line 99
If the file exists, merge new decisions into it.
- Distinctive: line 103
If a newer decision contradicts an older one:
- Distinctive: line 106
Add a note linking to the newer decision.
- Distinctive: line 107
Do not delete the old decision — it provides context.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Decision Memory
Creator: Neeraj Project: ClaudeSkills
Preserve important project decisions so that future AI sessions, new developers, or future you do not silently contradict previous choices.
Core principle: decisions made without recording them are decisions that will be unmade. Every project accumulates important choices — why PostgreSQL over MongoDB, why a monolith over microservices, why this folder structure. Without documentation, each new session starts from scratch and may "optimize" away intentional decisions.
This skill is a decision recording and retrieval tool — not a planning tool, not a documentation tool.
When to activate
Activate when:
- The user says "record this decision", "document why we chose X", "save this context".
- A significant architectural choice was just made in the conversation.
- The user wants to prevent future agents from contradicting current choices.
- Starting work on a project that has no decision records.
- The user says "remember this" or "make sure future sessions know about this".
Phase 1 — Search for existing decisions
Before creating anything, find what already exists:
- Check for
PROJECT_DECISIONS.md,docs/decisions/,ADR/,architecture/directories. - Read
README.md,CLAUDE.md,PROJECT_CONTEXT.mdfor embedded decisions. - Check git history for commit messages that describe decisions.
- Look for inline code comments that explain "why" a choice was made.
Never overwrite existing decisions without comparing.
Phase 2 — Identify decisions from code evidence
Inspect the project for decisions that are implied by the code but not documented:
- Framework choice — which framework, why (infer from code if not stated).
- Database choice — which database, which ORM.
- API conventions — REST vs GraphQL, naming conventions, versioning approach.
- UI conventions — component library, styling approach, state management.
- Security decisions — authentication approach, session handling.
- Dependency choices — key libraries chosen and alternatives that exist.
- Folder structure — organizational pattern (feature-based, layer-based, etc.).
- Testing approach — test framework, coverage expectations, E2E strategy.
For each, note the evidence (file, code, config) that supports the inference.
Phase 3 — Distinguish decision types
Classify each finding carefully:
- DECISION — an intentional choice with reasoning (e.g., "We chose X because Y").
- PREFERENCE — a style choice without strong reasoning (e.g., "Tabs over spaces").
- TEMPORARY DETAIL — a current state that may change (e.g., "Using v3.2.1").
- UNKNOWN — the choice exists but the reason is not evident.
Only record DECISIONS and significant PREFERENCES in the decision file. Do not record temporary details that will quickly become stale.
Phase 4 — Create or update PROJECT_DECISIONS.md
If the file does not exist, create it with this structure:
# Project Decisions
This document records important architectural and design decisions.
Future developers and AI agents should respect these choices unless
there is a clear reason to revisit them.
---
## [Decision Title]
**Decision:** [What was decided]
**Context:** [What situation prompted this choice]
**Reason:** [Why this choice was made]
**Alternatives Considered:** [What other options were evaluated]
**Consequences:** [What this choice implies for the project]
**Date:** [When the decision was made, if known]
**Status:** Accepted / Superseded / Under ReviewIf the file exists, merge new decisions into it. Do not overwrite existing content.
Phase 5 — Identify superseded decisions
If a newer decision contradicts an older one:
- Mark the older decision as Superseded.
- Add a note linking to the newer decision.
- Do not delete the old decision — it provides context.
Phase 6 — Note gaps
Explicitly list decisions that are missing — areas where the project has no documented choice:
- "No decision recorded for: error tracking approach"
- "No decision recorded for: deployment strategy"
This helps the user identify what should be documented.
What this skill does NOT do
- Does not make architectural decisions (that is
product-architect). - Does not document the codebase (that is
context-keeper). - Does not create general documentation (that is
documentation-engineer). - Does not record every tiny choice — only significant architectural/design decisions.
Final output
# Decisions Found
# Decisions Recorded
# Decisions Updated
# Superseded Decisions
# Missing Decisions (recommendations)Guardrails: never invent historical decisions; always ground decisions in code evidence; distinguish decisions from preferences from temporary details; do not overwrite existing decision records; merge new decisions into existing files; keep decisions concise and useful; and never claim a decision was made if the project does not support that fact.
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.

