feat(dev-preview): 10-minute offline quickstart + resume demo#124
Merged
Conversation
Add a no-API-key dev-preview path: root editable install, a skill smoke test, and a deterministic with/without-x.klickd task-resume demo that emits a scorecard. Governance rules are read live from the bundled x.klickd/coding skill. Public release stays v4.1; demo is explicitly not a model benchmark.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds a developer preview path so an external developer can see what x.klickd structured memory/skill context does in under 10 minutes, with no API key, no account, and no secrets. Scope is intentionally reduced to Python CLI + one demo, per the Day 2 brief.
pyproject.tomlmakespip install -e .work from a fresh clone, installing the same publishedklickdpackage whose source lives atpackages/pypi/klickd/(no code duplication, version stays 4.1.0).examples/dev-preview/hello_skill.py: loads a bundled starter skill and one of the 42 v4.1 candidate packs, hash-verified against the published manifest. No LLM.examples/dev-preview/run_demo.py: deterministic, offline simulation of an agent resuming an interrupted coding task two ways — bare prompt vs. carried memory + skill gates. The human-veto/verification-gate rules the x.klickd lane obeys are read live from the bundledx.klickd/codingskill (not hardcoded), and the script emitsresults/comparison_scorecard.md.examples/dev-preview/README.md: what it proves, what it does not prove (not a model benchmark, not native client support, no compliance claim), and how it relates to the internal supply chain without exposing any private detail.tests/test_dev_preview.py: asserts both scripts run clean, the scorecard is generated, the two lanes genuinely diverge, the scorecard governance matches the real skill, and no forbidden release/benchmark phrasing leaks in.docs/getting-started.md: new2cdev-preview section.Truth boundary (no mirage)
.internal-skillsinputs.Test plan
pip install -e .from repo root (fresh venv) → importsklickd4.1.0, 42 packs.python examples/dev-preview/hello_skill.py→ exit 0.python examples/dev-preview/run_demo.py→ exit 0, scorecard written, lanes diverge (risky actions 2 → 0).pytest tests/test_dev_preview.py→ 5 passed.docs/getting-started.md+examples/dev-previewfor forbidden release/benchmark claims and internal-path leakage → none.🤖 Generated by Computer