Skip to content

feat(slack): let bare readers inspect the bot’s recent run - #141

Merged
Shashankss1205 merged 2 commits into
mainfrom
fix/slack-recent-run
Oct 15, 2026
Merged

Shashankss1205 merged 2 commits into
mainfrom
fix/slack-recent-run

Conversation

@Shashankss1205

@Shashankss1205 Shashankss1205 commented Oct 15, 2026 •

Copy link
Copy Markdown
Collaborator

After a bot run, bare metrics, viz, replay and trace now select a recent trace and state exactly which path and run id were selected. Remember completed traces per channel in a thread-safe bot-local history; fall back to the newest trace under slack-runs/ in a new channel or after restart, and disclose that fallback.

Explicit arguments keep their current behavior. Resolved arguments pass the existing confinement/allowlist gate; empty history, corrupt traces and multi-run ambiguity produce actionable refusals. diff keeps its explicit comparison baseline. Channel history is process-local, as documented.

Add 17 regression checks, including the real demo stage0 → metrics subprocess flow, channel selection and symlink confinement; update the Slack cookbook and selected-test figure. The pre-integration behavior fails 13 new checks. Focused Slack/documentation checks: 151 passed. The final integrated tree through #143 passes 2,395 tests on both Python 3.12 and 3.13 locally; 13 skipped and 13 paid tests excluded. Ruff/diff checks and all 133 installed-wheel module imports pass. GitHub CI results are recorded in the review before merge.

No Slack credentials, network connection or paid model calls are used by the tests.

Fixes #49.

@Shashankss1205 Shashankss1205 left a comment

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Evaluation of 4073e86: bare readers disclose the inferred trace/run and prefer per-channel completed history, with an explicit workdir fallback. Typed reader arguments keep their existing behavior. Inferred arguments re-enter the admission gate; candidate paths are confined before reads, corrupt newest traces do not silently select an older answer, and multiple runs require an explicit choice. Channel history is process-local as documented.

Validation: 17 new checks pass, including demo stage0 followed by metrics through the real CLI, missing history, channel isolation, malformed traces and escaping symlinks. The focused Slack/documentation run passes 151 tests. Full integrated local suite: 2,388 passed, 13 skipped, 13 paid tests deselected. Ruff/diff checks and all 133 installed-wheel module imports pass. GitHub CI is pending; merge after #140 and successful checks.

No blocking code findings. diff correctly retains an explicit comparison baseline.

Copy link
Copy Markdown
Collaborator Author

Final merge check for 4073e86: all GitHub CI test jobs (Python 3.12, 3.13 and 3.14), lint, distribution validation and the live-test guard passed. #140 is merged; the remaining diff contains only the five Slack-related files. The reviewed head is unchanged, with no unresolved review threads or requested changes. Merging with ancestry preserved for #143.

@Shashankss1205
Shashankss1205 merged commit 7dec0f4 into main Oct 15, 2026
7 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

slack: metrics/viz/replay demand a path and run id the user cannot know — default to the bot's own last run

1 participant