amm-founding-circle / shep
AGENTS.md0 starsChanged 5 months agoMITView on GitHub
Read this before opening a merge request, adding a release gate, or treating anything here as customer-facing. Shep is internal research and development tooling. It is a governed terminal command deck that supervises agent sessions on a developer workstation. Its users are the engineers working in this repo. Shep is not a production application. It is not deployed to baremetal-production or baremetal-staging, it is not fronted by an Ingress, it serves no customers, and no customer data passes through it.…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 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 repository83 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- shep/CLAUDE.mdCLAUDE.md
- advisory-version-checkSkill
- aeo-llm-content-plannerSkill
- agency-command-centerSkill
- agency-morning-briefSkill
- agency-scorecardSkill
- agent-runbookSkill
- agent-usageSkill
- autonomy-budgetSkill
- backlink-pr-angle-generatorSkill
- brand-kit-from-urlSkill
- browser-automationSkill
- building-autonomous-agentsSkill
- capability-mapSkill
- case-study-engineSkill
- clickup-apiSkill
- client-dashboardSkill
- client-health-scoreSkill
- client-onboarding-osSkill
- cli-llm-routingSkill
- codebase-mapSkill
- command-from-your-phoneSkill
- competitor-content-gap-analysisSkill
- connection-monitorSkill
- content-brief-generatorSkill
- determinism-patternSkill
- dispatch-and-briefSkill
- division-operating-modelSkill
- dreamingSkill
- durable-stateSkill
- entity-topical-authority-mapperSkill
- first-real-jobSkill
- founding-circle-ai-securitySkill
- gbp-competitor-auditSkill
- goal-modeSkill
- host-your-agentSkill
- html-reportsSkill
- instinct-bridgeSkill
- evolveSkill
- learn-evalSkill
- learnSkill
- promoteSkill
- skill-createSkill
- internal-linking-auditorSkill
- keyword-deep-diveSkill
- ladder-auditSkill
- lanes-momentum-frameworkSkill
- launch-websiteSkill
- llm-citation-auditSkill
- migration-seo-paritySkill
- migration-visual-checkSkill
- model-gear-routerSkill
- multi-account-gatewaySkill
- multi-model-councilSkill
- night-shiftSkill
- novaSkill
- per-client-memory-routerSkill
- The first 60 shown.
The file
We split this AGENTS.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 3
Read this before opening a merge request, adding a release gate, or treating anything here as customer-facing.
- Very specific: line 53
Every change runs the full matrix before merge.
- Distinctive: line 68
If Shep gains a deployed surface, a customer, or a hosted endpoint, delete this exemption and restore the full evidence requirement before …
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
Shep — repository scope
Read this before opening a merge request, adding a release gate, or treating anything here as customer-facing.
What this repository is
Shep is internal research and development tooling. It is a governed terminal command deck that supervises agent sessions on a developer workstation. Its users are the engineers working in this repo.
What this repository is not
Shep is not a production application. It is not deployed to baremetal-production or baremetal-staging, it is not fronted by an Ingress, it serves no customers, and no customer data passes through it. There is no production surface to protect, no release train, and no on-call rotation.
What that means in practice
| Standard SearchAtlas process | Applies here? |
| Merge requests reviewed before merge | Yes |
| Tests + inventory pins green before merge | Yes — see ## Verify in the README |
| Linear issue link in the MR description | No — waived, see below |
| ClickUp Clip (QA recording) in the MR description | No — waived, see below |
| Deployment / rollback plan | No — nothing is deployed |
The MR evidence requirement is waived here
The company rule requires every merge request to carry a real Linear issue link and a real ClickUp Clip recording. That rule exists to prove that a change affecting customers was ticketed and demonstrated. This repository has no customer-facing surface, so the requirement does not apply to it. Waived explicitly by Project Owner on 2026-09-16.
This is a scope exemption, not a lowered bar. It does not licence a placeholder, dummy, or borrowed URL in any repository, here or anywhere else — a fake evidence link is a false claim and is never acceptable. If a change in this repo does happen to have a real ticket, link it.
The verification gates below are not waived and are the evidence for changes here.
What is currently in flight
docs/open-work.md is the list of live work items. Check it before starting something, so two efforts do not collide. It is maintained by hand — if you open, finish, or abandon an item, edit that file in the same merge request.
Verification is still mandatory
Every change runs the full matrix before merge. Use python3.11 — the host
python3 may be 3.9, and Shep requires >= 3.10.
python3.11 -m pytest -q
python3.11 tools/verify_test_inventory.py
python3.11 -m compileall -q scripts tests
ruff check scripts testsThe pinned node inventory in tests/pinned-node-ids.txt is order-sensitive:
regenerate it in pytest collection order and rerun the verifier.
If this ever changes
If Shep gains a deployed surface, a customer, or a hosted endpoint, delete this exemption and restore the full evidence requirement before the first deploy.
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.