Agent-FleetOps
AGENTS.md34 starsChanged 37 days agoMITView on GitHub
Public CI does not replace the private prepublication checks for secrets, personal data and provenance. Keep private identity and provenance inputs out of Actions. Dependency lower bounds allow CI resolution to vary between runs. A green run describes its resolved environment; it does not establish reproducibility through a dependency lock. After the test job on 3857c81 reached GitHub's six-hour cancellation limit while its guard step was running (20,682 s), the job is bounded at 100 minutes and its hermetic and…
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 repository31 other files it gives its agents.
- tui/AGENTS.mdAGENTS.md
- actionability-passSkill
- agent-memory-opsSkill
- blocked-page-retrievalSkill
- brain-consistency-auditorSkill
- brainstorm-panelSkill
- concurrent-push-integrationSkill
- curation-auditSkill
- decomposed-local-researchSkill
- dual-model-reconciliationSkill
- eval-integritySkill
- file-organizationSkill
- fleet-model-routingSkill
- fleet-operational-runbookSkill
- generate-review-fix-loopSkill
- guard-target-correctnessSkill
- honesty-stop-gateSkill
- large-file-cartographerSkill
- local-lane-build-loopSkill
- local-model-onboardingSkill
- model-routing-tableSkill
- multi-agent-code-workflowSkill
- peer-drop-watcherSkill
- protected-function-guardSkill
- research-dispatchSkill
- research-verificationSkill
- research-workflowSkill
- shared-brain-bookkeepingSkill
- should-weSkill
- task-dependency-sequencingSkill
- verify-running-buildSkill
The file
We split this AGENTS.md into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 4
Run the TUI and publication guard checks for main pushes and pull requests.
- Very specific: line 7
Keep workflow token permissions at contents: read, disable checkout credential persistence, and pin actions to verified upstream commit …
≈ “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
GitHub automation
Purpose
Run the TUI and publication guard checks for main pushes and pull requests.
Local Contracts
Keep workflow token permissions at contents: read, disable checkout credential persistence, and pin actions to verified upstream commit hashes. Full-history checkout is required by the guard checks.
Work Guidance
Preserve triggers, job names, Python versions and test commands when changing workflow security settings. Review upstream action pins before updating them.
Public CI does not replace the private prepublication checks for secrets, personal data and provenance. Keep private identity and provenance inputs out of Actions.
Dependency lower bounds allow CI resolution to vary between runs. A green run describes its resolved environment; it does not establish reproducibility through a dependency lock.
After the test job on 3857c81 reached GitHub's six-hour cancellation limit while its guard step was running (20,682 s), the job is bounded at 100 minutes and its hermetic and guard steps at 45 minutes each. The same guard step on ee18a1d took 584 s on the hosted runner and 423 s on a local machine (1.38×). On that local machine the bounded suite's whole guard step took 676 s, of which its pytest guard/tests/ took 670 s, so the 45-minute guard deadline leaves about three times the expected hosted time. The hermetic step took 970 s on 3857c81, so its 45-minute deadline leaves about 2.8 times that. The python-floor and ref-gate jobs are bounded at 10 minutes each. Preserve these deadlines when editing the workflow.
Verification
Parse workflow YAML and review the exact diff. Confirm both test and ref-gate pass in GitHub Actions for the published commit; local syntax checks do not prove hosted execution.
The python-floor job uses Python 3.11 to ast-parse every tracked tui/**/*.py file (including files not imported during collection), failing on any SyntaxError, then installs TUI requirements and collects the suite. Confirm its hosted result for the published commit alongside test and ref-gate; local floor checks do not prove hosted execution.
Child DOX Index
workflows/tui-tests.yml defines the checks.
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.