save-knowledge
Skill1 starsChanged 7 months agoMITView on GitHub
Use when a decision, discovery, plan outcome, or important finding needs to be saved for future reference. Routes to kb inbox (if available) AND appends to episodic memory for kb-dream. Falls back to docs/decisions/ when kb is not installed.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- teachskillofskills-ai/DigitalMarketingPro-techshutheir own version, first seen Oct 11, 2026
- indranilbanerjee/digital-marketing-protheir own version, first seen Feb 12, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- indranilbanerjee/digital-marketing-pro★ 934same name
- teachskillofskills-ai/DigitalMarketingPro-techshu★ 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 repository8 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- autoresearchSkill
- chrome-devtoolsSkill
- experiment-runnerSkill
- playwright-browserSkill
- pr-reviewSkill
- skill-creatorSkill
- sp-plan-discoverySkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 0 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 18
NEVER use kb add "text" — it will fail with "file not found"
- Very specific: line 22
When kb is available, every save goes to two places:
- Very specific: line 68
When saving knowledge with kb available:
- Very specific: line 74
When kb is NOT available:
- Very specific: line 75
Create docs/decisions/YYYY-MM-DD-<title>.md in the current project
- Very specific: line 93
When kb is NOT available, create docs/decisions/YYYY-MM-DD-<title>.md:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
save-knowledge
Save decisions and discoveries to the right place, automatically.
Priority Order
- If
kbcommand exists →kb "<structured content>"(sends text directly to inbox as a timestamped.mdfile) - Else → create
docs/decisions/YYYY-MM-DD-<title>.mdin the current project directory
IMPORTANT — kb CLI usage:
kb "text"→ saves text to inbox (CORRECT for saving knowledge)kb add <file>→ copies an existing FILE to inbox (only for files, NOT text)- NEVER use
kb add "text"— it will fail with "file not found"
Two Save Targets
When kb is available, every save goes to two places:
1. Inbox (for classification by kb process)
kb "Title: <descriptive title>
Decision: [what was decided]
Why: [the reason]
Context: [project, component]"2. Episodic Memory (for consolidation by kb-dream)
Append claims to $KB_VAULT/memory/YYYY-MM-DD.md. This file accumulates
claims throughout the day. kb-dream reads these to build MEMORY.md and
promote high-confidence findings to raw/memory-derived/.
# Append to today's memory file — create if it doesn't exist
MEMORY_FILE="$KB_VAULT/memory/$(date +%Y-%m-%d).md"Each claim uses this YAML format:
- claim: "Short factual statement of what was decided or discovered"
memory_type: decision | technical-finding | user-preference | user-rule
confidence: low | medium | high
context: "Project or area this relates to"memory_type guide:
| Type | When to use | Example |
decision | A choice between alternatives was made | "Chose Regional LB over Global LB for webhook traffic" |
technical-finding | Something non-obvious was discovered or validated | "Cloud Armor policy is active on the LB backend service" |
user-preference | User expressed a preference for how things should work | "User wants gemini-3-flash-preview only" |
user-rule | User set a rule that must always be followed | "Never log webhook payloads" |
confidence guide:
| Level | Criteria |
high | Verified by evidence (logs, config, tests, user confirmation) |
medium | Reasonable inference from available data |
low | Speculation or unverified assumption |
Workflow
When saving knowledge with kb available:
- Send structured content to inbox:
kb "Title: ..." - Append claim(s) to
$KB_VAULT/memory/$(date +%Y-%m-%d).md - Both happen every time — never skip memory
When kb is NOT available:
- Create
docs/decisions/YYYY-MM-DD-<title>.mdin the current project - Memory is not written (no vault access)
What to Save
Save when you encounter:
- A decision between multiple approaches (what was chosen and why)
- A discovery that wasn't obvious from the code or docs
- A constraint that will affect future work
- The outcome of an experiment or investigation
What to Include
Structure every save with a clear title line followed by structured content.
The content will be processed later by kb process which classifies and archives it.
Markdown Fallback Format
When kb is NOT available, create docs/decisions/YYYY-MM-DD-<title>.md:
# [Title]
**Date:** YYYY-MM-DD
## Decision
[What was decided]
## Why
[Rationale and constraints]
## Alternatives Discarded
- [Option A]: [why rejected]
- [Option B]: [why rejected]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.

