OwnPilot
Copilot instructions426 starsChanged 3 months agoMITView on GitHub
This project uses DFMT to keep tool output from flooding the context window and to preserve session state across compactions. When working in this project, follow these rules. Prefer DFMT's MCP tools over native ones: Every dfmt_* call MUST pass an intent parameter — a short phrase describing what you need from the output (e.g. "failing tests", "error message", "imports"). Without intent the tool returns raw bytes and the token savings are lost. On DFMT failure, report it to the…
We split this copilot-instructions.md into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
Prefer DFMT's MCP tools over native ones:
- Very specific: line 23
Every dfmt_* call MUST pass an intent parameter — a short phrase describing what you need from the output (e.g.
- Distinctive: line 33
If the native tool is also denied (permission rule, sandbox refusal), stop and ask the user.
≈ “Common” is an estimate: at least 1 in 5 of the 3,669 Copilot instructions 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
<!-- dfmt:v1 begin -->Context Discipline
This project uses DFMT to keep tool output from flooding the context window and to preserve session state across compactions. When working in this project, follow these rules.
Tool preferences
Prefer DFMT's MCP tools over native ones:
| Native | DFMT replacement | intent required? |
Bash | dfmt_exec | yes |
Read | dfmt_read | yes |
WebFetch | dfmt_fetch | yes |
Glob | dfmt_glob | yes |
Grep | dfmt_grep | yes |
Edit | dfmt_edit | n/a |
Write | dfmt_write | n/a |
Every dfmt_* call MUST pass an intent parameter — a short phrase
describing what you need from the output (e.g. "failing tests",
"error message", "imports"). Without intent the tool returns raw
bytes and the token savings are lost.
On DFMT failure, report it to the user (one short line — which call,
what error) and then fall back to the native tool so the session is
not blocked. The ban is on silent fallback — every switch must be
announced. After a fallback, drop a brief dfmt_remember note tagged
gap when practical, so the journal records that a call was bypassed.
If the native tool is also denied (permission rule, sandbox refusal),
stop and ask the user; do not retry blindly.
Session memory
DFMT tracks tool calls automatically. After substantive decisions or
findings, call dfmt_remember with descriptive tags (decision,
finding, summary) so future sessions can recall the context after
compaction.
When native tools are acceptable
Native Bash and Read are acceptable for outputs you know are small
(< 2 KB) and will not be referenced again. For everything else, DFMT
tools are preferred.
<!-- dfmt:v1 end -->Where files are headingNew Copilot instructions files have more rules about branches and pull requests, and fewer about code style.
Of the Copilot instructions files we read, the 2,297 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository3 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
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.