Orchestra memory
CLI to validate, scaffold, and install Agent Skills for Cursor and agentskills.io-compatible agents
npx -y skills add luigipascal/skill-forge --skill orchestra-memoryAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Queries Orchestra shared memory at localhost:4317 before coding when context may exist in prior chats, synced repos, or ingested threads. Use when the user mentions Orchestra memory, Untitled threads, recall, or cross-project context.
SKILL.md
1.5 KB, 310 tokens by cl100k_base, as published. Nobody here has run it
Orchestra Memory
Orchestra runs locally (default http://localhost:4317). Use it to avoid re-discovering decisions already captured in memory.
When to recall
- User references a prior ChatGPT/NotebookLM thread by URL or nickname (e.g. "Untitled-1")
- Starting work on a repo that is in Orchestra sync paths
- Cross-project decisions (OrbaLang, Forum, Indie, Gustavo, Orchestra itself)
Quick checks
# Health + stats
Invoke-RestMethod http://localhost:4317/api/health
# Search by keyword
Invoke-RestMethod "http://localhost:4317/api/search?q=YOUR_QUERY&limit=10"
# Recall block for agent context
Invoke-RestMethod -Method POST -Uri http://localhost:4317/api/recall `
-ContentType "application/json" `
-Body '{"query":"YOUR_QUERY","limit":8}'
# Fetch conversation by ID (from search hits)
Invoke-RestMethod "http://localhost:4317/api/conversations/CONVERSATION_ID"
Match ChatGPT URLs
If the user gives https://chatgpt.com/c/<id>, search conversations:
source_idoften contains/c/<id>- Title may differ from "Untitled-1" — match on URL fragment, not title
After recall
Summarize relevant memory for the user, note conflicts with current task, then proceed with implementation.
Gives 0 of the 12 instructions most memory context skills give in 310 tokens
Counted across 674 of the 847 authors here whose files we hold, read 2026-08-06
- inform the user when setup is completein 21 of 674, across 6 files
- confirm the draft with the user before writingin 21 of 674, across 6 files
- update the agent skills block in place if it existsin 21 of 674, across 6 files
- present findings to the userin 20 of 674, across 5 files
- write the three docs files from seed templatesin 20 of 674, across 5 files
- ask the user about each decision one at a timein 19 of 674, across 4 files
- edit CLAUDE.md if it existsin 18 of 674, across 3 files
- explore current repo statein 18 of 674, across 3 files
- do not overwrite user edits to surrounding sectionsin 18 of 674, across 3 files
- back up the original file before overwritingin 16 of 674, across 8 files
- keep the memory index under 200 linesin 15 of 674
- Provide actionable steps and verificationin 13 of 674, across 2 files
Said here and by no other author read
- query Orchestra memory before coding
- check service health and stats first
- fetch recall blocks via POST request
- fetch conversations by ID
- match ChatGPT URLs on URL fragment not title
- note conflicts with current task
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.