llm-context-change
Skill52 starsChanged 47 days agoApache-2.0View on GitHub
Checklist after changing what the LLM receives (skforecast_ai/llm/context.py, llm/prompts.py, the rendered explanations in recommendation/explanation.py, execution/backtesting_runner.py or execution/comparison.py). Regenerates and reviews the golden contexts and prepares the real-model check that the user launches.
In this repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- handoffSkill
- open-prSkill
- release-bumpSkill
- release-noteSkill
- sync-skforecast-assetsSkill
- verifySkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 1 is distinctive, and 4 are 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.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
Run the golden test to see what changed: PY -m pytest tests/tests_llm/test_golden_context.py -q (PY as in /verify).
- Very specific: line 11
If the difference is intended, regenerate: PY tools/ai/update_golden_contexts.py
- Distinctive: line 15
Check that no raw dataset values leaked into a context (privacy by default).
- Very specific: line 21
Check the token budget tests: PY -m pytest tests/tests_llm -q.
- Very specific: line 26
If the answers of ask() change in a way a user notices, add a release note with /release-note.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
LLM context change
- Run the golden test to see what changed:
PY -m pytest tests/tests_llm/test_golden_context.py -q(PYas in/verify). - If the difference is intended, regenerate:
PY tools/ai/update_golden_contexts.py - Review
git diff tests/tests_llm/golden/and summarize it for the user: what the model now sees that it did not, and what it no longer sees. Check that no raw dataset values leaked into a context (privacy by default); only summary statistics and values owned by a result are allowed. - Build the contexts without calling a model:
PY tools/ai/check_ask_context.py --dry-runand skim the report for truncation or empty sections. - Check the token budget tests:
PY -m pytest tests/tests_llm -q. - Tell the user that this change needs a real-model run of
tools/ai/check_ask_context.pybefore the release. It costs money, so the user launches it (a hook blocks it without--dry-run); the reviewed report is kept as described intools/ai/ask_context_reports/README.md. - If the answers of
ask()change in a way a user notices, add a release note with/release-note.
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.

