- Home
- Skills
- Workflow Automation
- StreamTeem Project Memory
Works with the AI Agents you already use
StreamTeem Project Memory
Five skills that give your agent a lasting memory of a project, in plain markdown files: facts it should know and rules it must follow. Loads what applies when work starts and re-checks rules before commit, push, or PR.
$1.99
StreamTeem Project Memory
Example session with this skill installed
I am about to push these changes to the remote branch. Check the project memory for any rules that apply to this action.
- Read your context and instructions
- Compiled the streamteem project memory
I have checked the project memory for the git:push trigger within the developer workflow. One relevant rule was found and verified.
Pre-push Memory Check
| Entry Type | Title | Workflow | Trigger | Status |
|---|---|---|---|---|
| Rule | Integration tests hit real DB | developer | git:push | Requirement Met |
| Fact | CI/CD Pipeline URL | all | N/A | Reference Only |
Analysis:
The rule Integration tests hit real DB requires that any push to a remote branch must be preceded by a local execution of the integration suite against the staging container. I have confirmed that the pytest --integration command was executed successfully in the previous step of this session.
Next steps
- Proceed with
git push origin feature-branch. - Monitor the
pr:opentriggers if you intend to create a pull request immediately after. - Use
rememberif any new deployment constraints were discovered during this push.
Connects securely to your tools. The creator never sees your data.
About this skill
Context drift and forgotten architectural constraints lead to technical debt and deployment failures. The Streamteem Project Memory skill package provides a persistent, workflow-aware knowledge system that lives directly in your repository. It ensures that critical facts and procedural rules are loaded, followed, and evolved throughout the development lifecycle.
What it does
- Load context automatically reads stored facts and rules at the start of a session or task to ensure constraints are in force.
- Capture knowledge classifies new observations as facts or prescriptive rules at either a specific work-item or global expert-role scope.
- Enforce triggers checks specific rules immediately before critical actions like
git:push,pr:open, or writing SQL. - Integrate documentation parses external notes, runbooks, and code comments to batch-process them into categorized memory files.
- Optimize memory runs Python-based analysis to detect file bloat, duplicates, and missing metadata to maintain high context performance.
- Promote learnings elevates successful patterns from individual tasks to the project-wide expert memory.
How it works
- Initialize by defining a memory folder to store Markdown-based fact and rule files.
- Execute tasks while the agent monitors for new patterns or follows existing rules tied to specific triggers.
- Commit memory by saving captured knowledge into workflow-split files like
facts-developer.mdorrules-incident-response.md. - Maintain health by running the optimization script to prune duplicates and fix indexing.
Frameworks & tools
This skill manages Markdown files within a Git repository. It uses a bundled Python 3 script for memory analysis and supports integration with standard CLI tools for version control and database operations.
Why this beats prompting it yourself
Standard prompts lose context as conversations grow long, and long system instructions waste tokens on irrelevant data. This system uses a workflow-split architecture that only loads the specific facts and rules relevant to your current task, reducing noise and preventing "context hallucination."
Use cases
- Enforcing database deployment order rules before any SQL is written.
- Onboarding new developers by loading architecture facts and team communication protocols.
- Capturing incident response post-mortems as permanent rules for the SRE workflow.
- Auditing project memory to remove stale environment URLs and redundant documentation.
Known limitations
Requires a specific Markdown file structure to be maintained in the project root. The optimization script requires a local Python 3 environment.
Requires local Python 3 for optimization script. Must follow specific Markdown file naming conventions. Scope is limited to the local project repository.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 4 days ago
Requires
Permissions
Read/write for facts and rules Markdown. Shell only when running the bundled memory-optimize Python script.